CN108920513A - A kind of multimedia data processing method, device and electronic equipment - Google Patents
A kind of multimedia data processing method, device and electronic equipment Download PDFInfo
- Publication number
- CN108920513A CN108920513A CN201810549255.1A CN201810549255A CN108920513A CN 108920513 A CN108920513 A CN 108920513A CN 201810549255 A CN201810549255 A CN 201810549255A CN 108920513 A CN108920513 A CN 108920513A
- Authority
- CN
- China
- Prior art keywords
- audio
- target
- information
- audio section
- section
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 18
- 238000004458 analytical method Methods 0.000 claims abstract description 167
- 230000002123 temporal effect Effects 0.000 claims abstract description 113
- 230000006399 behavior Effects 0.000 claims description 110
- 230000003542 behavioural effect Effects 0.000 claims description 98
- 238000013145 classification model Methods 0.000 claims description 73
- 238000000034 method Methods 0.000 claims description 44
- 238000005520 cutting process Methods 0.000 claims description 27
- 238000003860 storage Methods 0.000 claims description 14
- 238000006243 chemical reaction Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 description 23
- 238000010586 diagram Methods 0.000 description 15
- 230000037007 arousal Effects 0.000 description 6
- 230000002996 emotional effect Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 230000029058 respiratory gaseous exchange Effects 0.000 description 2
- 238000012552 review Methods 0.000 description 2
- 241001633942 Dais Species 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000007596 consolidation process Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 235000015096 spirit Nutrition 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Educational Administration (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Technology (AREA)
- Multimedia (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Acoustics & Sound (AREA)
- Human Computer Interaction (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The embodiment of the invention discloses a kind of multimedia data processing methods, including：Obtain the first audio-frequency information and the second audio-frequency information belonged in classroom environment；The audio section for belonging to teacher's type is searched in the first audio-frequency information, as the first audio section of target, and target second audio section is searched in the second audio-frequency information；Target second audio section temporal information corresponding with the first audio section of target does not overlap；The first audio section of target is converted into the first text information of target, the first type of teaching information corresponding with the first audio section of target is identified according to the first text information of target；According to the first type of teaching information, the corresponding temporal information of the first audio section of target, the corresponding temporal information of target second audio section, the corresponding audio types of target second audio section and audio frequency characteristics, the key link information in classroom environment is counted.Using the present invention, the analysis efficiency and precision of analysis of Classroom instruction quality can be improved.
Description
Technical field
The present invention relates to field of computer technology more particularly to a kind of multimedia data processing method, device and electronics to set
It is standby.
Background technique
Classroom instruction is the instructional mode expansion with " teacher be leading, based on student ".Teacher in teaching process
The degree of participation of teaching quality and student will have a direct impact on student to the degree of understanding of theoretical knowledge and professional skill.Therefore it is
Improve the quality of teaching, the process of teaching effectively managed it is most important, by teaching process teacher and
Raw performance exercises supervision and scientifically analyzes, and is just able to achieve teaching process and fully and effectively manages.
In order to comprehensively analyze classroom, the existing quality surveillance to teaching classroom is based on manual oversight, is existed by supervisor
Give lessons situation and the study situation of student of teaching classroom on-site supervision teacher.The classroom situation of monitoring is analyzed again
And statistics, to evaluate Classroom instruction quality, but Classroom instruction quality is analyzed by way of manual oversight, there can be subjectivity
And unstability, it cannot accomplish accurately to analyze Classroom instruction quality.
It is above-mentioned as it can be seen that manual analysis Classroom instruction quality inefficiency, can not accomplish efficiently, reasonably analyze Classroom instruction quality.
Summary of the invention
The technical problem to be solved by the embodiment of the invention is that provide a kind of multimedia data processing method, device and
The analysis efficiency and precision of analysis of Classroom instruction quality can be improved in electronic equipment.
First aspect of the embodiment of the present invention provides a kind of multimedia data processing method, including：
Obtain the first audio-frequency information and the second audio-frequency information belonged in classroom environment；First audio-frequency information and institute
Stating the second audio-frequency information is that the radio reception device for being located at different location by two respectively collects；
The audio section for belonging to teacher's type is searched in first audio-frequency information, as the first audio section of target, and
Target second audio section is searched in second audio-frequency information；The target second audio section and first audio section of target
Corresponding temporal information does not overlap；
First audio section of target is converted into the first text information of target, according to first text information of target
Identify the first type of teaching information corresponding with first audio section of target；
According to the first type of teaching information, the corresponding temporal information of the first audio section of the target, the target
The corresponding temporal information of two audio sections, the corresponding audio types of the target second audio section and audio frequency characteristics, count the class
Key link information in hall environment；The key link information includes teacher's behavioural analysis result and students ' behavior analysis knot
Fruit.
Wherein, described that the audio section for belonging to teacher's type is searched in first audio-frequency information, as the first sound of target
Frequency range, including：
According to audio frequency characteristics by the first audio-frequency information cutting be multiple the first audio sections of unit；Each unit first
Audio section is the audio section comprising continuous sound or is mute audio section；
Each first audio section of unit is input in audio classification model, respectively identification and each unit
The matched audio types of first audio section；
It is the first audio section of unit of teacher's type by the audio types in first audio-frequency information,
It is determined as first audio section of target.
Wherein, the lookup target second audio section in second audio-frequency information, including：
According to audio frequency characteristics by the second audio-frequency information cutting be multiple the second audio sections of unit；Each the second audio of unit
Section is the audio section comprising continuous sound or is mute audio section；
Each second audio section of unit is input in the audio classification model, is identified respectively and described each
The matched audio types of the second audio section of unit；
It is not the second audio section of unit of teacher's type by the audio types in second audio-frequency information, as
Audio section to be adjusted；
If the temporal information of audio section to be adjusted and the temporal information of first audio section of target be not be overlapped, it is determined that
If being target second audio section for the audio section to be adjusted；
It, will be described if the temporal information of audio section to be adjusted is Chong Die with the temporal information of first audio section of target
The not corresponding audio section of lap in audio section to be adjusted, is determined as the target second audio section.
Wherein, described according to first text information of target identification and first audio section of target corresponding first
Type of teaching information, including：
First text information of target is inputted into the first textual classification model, is obtained and first audio section of target
Matched teaching means information；
First text information of target is inputted into the second textual classification model, is obtained and first audio section of target
Matched content of courses structure；
By the corresponding teaching means information of first audio section of target and the corresponding teaching of the first audio section of the target
Content structure is determined as the first type of teaching information corresponding with first audio section of target.
Wherein, further include：
Target second audio section by audio types for single student's type or more people student's types is converted to target second
Text information, by target first text information and the target second text envelope adjacent with second text information of target
Breath is input in first textual classification model, is obtained and the matched teaching means information of the target second audio section；
By target first text information and the target second text information adjacent with second text information of target
It is input in second textual classification model, obtains and the matched content of courses structure of the target second audio section；
By the corresponding teaching means information of the target second audio section and the corresponding teaching of the target second audio section
Content structure is determined as the second type of teaching information corresponding with the target second audio section.
Wherein, further include：
Obtain the index model to match with the classroom environment；The index model is the classroom constructed by teaching notes
In knowledge-point models；
The knowledge point text in the index model is extracted, in first text information of target, determination is known with described
Know the object time stamp of the text information of point text matches, and the position of the object time stamp in first audio-frequency information
Add Keyword Tag；Text in the Keyword Tag is identical as the knowledge point text；
When getting searching keyword, retrieved in first audio-frequency information matched with the searching keyword
Keyword Tag, as target keywords label；
In first audio-frequency information, audio-frequency information corresponding with the timestamp of the target keywords label is exported,
And determine the curriculum information of audio-frequency information corresponding with the target keywords label.
Wherein, it is described according to the first type of teaching information, the corresponding temporal information of the first audio section of the target,
The corresponding temporal information of the target second audio section, the corresponding audio types of the target second audio section and audio frequency characteristics,
The key link information in the classroom environment is counted, including：
According to the first type of teaching information, the corresponding temporal information of the first audio section of the target, teacher's row is generated
To analyze result；
According to the second type of teaching information, the corresponding temporal information of the target second audio section, the target
The corresponding audio types of two audio sections and audio frequency characteristics generate students ' behavior and analyze result；
Teacher's behavioural analysis result and students ' behavior analysis result are determined as the pass in the classroom environment
Key link information.
It is wherein, described according to the first type of teaching information, the corresponding temporal information of the first audio section of the target,
Teacher's behavioural analysis is generated as a result, including：
According in the first type of teaching information teaching means information and the first audio section of the target it is corresponding
Temporal information counts the total duration with the first audio section of target of identical teaching means information, obtains each teaching means letter
Cease corresponding the first parameter of teacher's behavioural analysis；
According to the teaching hand content structure and the first audio section of target correspondence in the first type of teaching information
Temporal information, count have identical content of courses structure the first audio section of target total duration, obtain each content of courses
Corresponding the second parameter of teacher's behavioural analysis of structure；
According to the teaching means information in the first type of teaching information, counting has identical teaching means information
The quantity of the first segment of target obtains the corresponding teacher's behavioural analysis third parameter of each teaching means information；
According to the content of courses structure in the first type of teaching information, counting has identical content of courses structure
The quantity of the first segment of target obtains corresponding the 4th parameter of teacher's behavioural analysis of each content of courses structure；
According to first parameter of teacher's behavioural analysis, the second parameter of teacher's behavioural analysis, teacher's behavioural analysis third ginseng
Several, described the 4th parameter of teacher's behavioural analysis, generates teacher's behavioural analysis result.
Wherein, it is described according to the second type of teaching information, the corresponding temporal information of the target second audio section,
The corresponding audio types of the target second audio section and audio frequency characteristics generate students ' behavior analysis as a result, including：
The quantity with the target second audio section of identical audio frequency characteristics and audio types is counted, as students ' behavior point
Analyse the first parameter；
According in the second type of teaching information teaching means information and the target second audio section it is corresponding
Temporal information counts the total duration with the target second audio section of identical teaching means information, obtains each teaching means letter
It ceases corresponding students ' behavior and analyzes the second parameter；
According in the second type of teaching information content of courses structure and the target second audio section it is corresponding
Temporal information counts the total duration with the target second audio section of identical content of courses structure, obtains each content of courses knot
The corresponding students ' behavior of structure analyzes third parameter；
According to the teaching means information in the second type of teaching information, counting has identical teaching means information
The quantity of the second segment of target obtains the corresponding students ' behavior of each teaching means information and analyzes the 4th parameter；
According to the content of courses structure in the second type of teaching information, counting has identical content of courses structure
The quantity of the second segment of target obtains the corresponding students ' behavior of each content of courses structure and analyzes the 5th parameter；
The first parameter is analyzed according to the students ' behavior, the students ' behavior analyzes the second parameter, the students ' behavior point
It analyses third parameter, the 4th parameter of students ' behavior analysis, the students ' behavior and analyzes the 5th parameter, generate student's row
To analyze result.
Second aspect of the embodiment of the present invention provides a kind of apparatus for processing multimedia data, including：
First obtains module, for obtaining the first audio-frequency information and the second audio-frequency information that belong in classroom environment；It is described
First audio-frequency information and second audio-frequency information are that the radio reception device for being located at different location by two respectively collects；
First searching module, for searching the audio section for belonging to teacher's type in first audio-frequency information, as mesh
Mark the first audio section；
Second searching module, for searching target second audio section in second audio-frequency information；The target second
Audio section temporal information corresponding with first audio section of target does not overlap；
Conversion module, for first audio section of target to be converted to the first text information of target；
Identification module, for according to target text information identification and first audio section of target corresponding first
Type of teaching information；
Statistical module, for being believed according to the first type of teaching information, the first audio section of the target corresponding time
Breath, the corresponding temporal information of the target second audio section, the corresponding audio types of the target second audio section and audio are special
Sign, counts the key link information in the classroom environment；The key link information includes teacher's behavioural analysis result and
Raw behavioural analysis result.
Wherein, first searching module, including：
First cutting unit is used to according to audio frequency characteristics be multiple the first sounds of unit by the first audio-frequency information cutting
Frequency range；Each the first audio section of unit is the audio section comprising continuous sound or is mute audio section；
First recognition unit is known respectively for each first audio section of unit to be input in audio classification model
Not with the matched audio types of each first audio section of unit；
First determination unit, for being teacher's type by the audio types in first audio-frequency information
The first audio section of unit is determined as first audio section of target.
Wherein, second searching module, including：
Second cutting unit is used to according to audio frequency characteristics be multiple the second audio sections of unit by the second audio-frequency information cutting；
Each the second audio section of unit is the audio section comprising continuous sound or is mute audio section；
Second recognition unit, for each second audio section of unit to be input in the audio classification model, point
It Shi Bie not be with the matched audio types of each second audio section of unit；
Second determination unit, for not being the list of teacher's type by the audio types in second audio-frequency information
The second audio section of position, as audio section to be adjusted；
Second determination unit, if be also used to audio section to be adjusted temporal information and first audio section of target
Temporal information be not overlapped, it is determined that if being target second audio section for the audio section to be adjusted；
Second determination unit, if be also used to audio section to be adjusted temporal information and first audio section of target
Temporal information overlapping be determined as the target the then by the not corresponding audio section of lap in the audio section to be adjusted
Two audio sections.
Wherein, the identification module, including：
Acquiring unit obtains and the mesh for first text information of target to be inputted the first textual classification model
Mark the matched teaching means information of the first audio section；
The acquiring unit, for first text information of target to be inputted the second textual classification model, acquisition and institute
State the matched content of courses structure of the first audio section of target；
Third determination unit, for by the corresponding teaching means information of the first audio section of target and the target the
The corresponding content of courses structure determination of one audio section is the first type of teaching information corresponding with first audio section of target.
Wherein, further include：
The conversion module is also used to audio types be single student's type or the target second of more people student's types
Audio section is converted to the second text information of target, by target first text information adjacent with second text information of target
It is input in first textual classification model, obtains and the target second audio section with second text information of target
Matched teaching means information；
First obtains module, be also used to target first text information adjacent with second text information of target and
Second text information of target is input in second textual classification model, is obtained and the target second audio section
The content of courses structure matched；
Determining module is used for the corresponding teaching means information of the target second audio section and the second sound of the target
The corresponding content of courses structure determination of frequency range is the second type of teaching information corresponding with the target second audio section.
Wherein, further include：
Second obtains module, for obtaining the index model to match with the classroom environment；The index model is logical
Cross the knowledge-point models in the classroom of teaching notes building；
Adding module, for extracting the knowledge point text in the index model, in first text information of target
In, the determining object time with the text information of the knowledge point text matches stabs, and the mesh in first audio-frequency information
Add Keyword Tag in the position for marking timestamp；Text in the Keyword Tag is identical as the knowledge braille text；
Retrieval module, for being retrieved and the inquiry in first audio-frequency information when getting searching keyword
The Keyword Tag of Keywords matching, as target keywords label；
Output module, for exporting the timestamp pair with the target keywords label in first audio-frequency information
The audio-frequency information answered, and determine the curriculum information of audio-frequency information corresponding with the target keywords label.
Wherein, the statistical module, including：
First generation unit, when being used for corresponding according to the first type of teaching information, the first audio section of the target
Between information, generate teacher's behavioural analysis result；
Second generation unit, when being used for corresponding according to the second type of teaching information, the target second audio section
Between information, the corresponding audio types of the target second audio section and audio frequency characteristics, generate students ' behavior analyze result；
4th determination unit, for teacher's behavioural analysis result and students ' behavior analysis result to be determined as
Key link information in the classroom environment.
Wherein, first generation unit, including：
First statistics subelement, for according in the first type of teaching information teaching means information and the mesh
The corresponding temporal information of the first audio section is marked, the total duration with the first audio section of target of identical teaching means information is counted,
Obtain corresponding the first parameter of teacher's behavioural analysis of each teaching means information；
The first statistics subelement, is also used to according to the teaching hand content structure in the first type of teaching information,
Temporal information corresponding with first audio section of target counts first audio section of target with identical content of courses structure
Total duration, obtain corresponding the second parameter of teacher's behavioural analysis of each content of courses structure；
The first statistics subelement, is also used to according to the teaching means information in the first type of teaching information, system
The quantity with the first segment of target of identical teaching means information is counted, it is corresponding old to obtain each teaching means information
Teacher's behavioural analysis third parameter；
The first statistics subelement, is also used to according to the content of courses structure in the first type of teaching information, system
The quantity with the first segment of target of identical content of courses structure is counted, it is corresponding old to obtain each content of courses structure
The 4th parameter of teacher's behavioural analysis；
First generates subelement, for being joined according to first parameter of teacher's behavioural analysis, teacher's behavioural analysis second
Number, teacher's behavioural analysis third parameter, the 4th parameter of teacher's behavioural analysis, generate teacher's behavioural analysis result.
Wherein, second generation unit, including：
Second statistics subelement, for counting the target second audio section with identical audio frequency characteristics and audio types
Quantity analyzes the first parameter as students ' behavior；
The second statistics subelement, is also used to according to the teaching means information in the second type of teaching information, and
The corresponding temporal information of the target second audio section, counts the target second audio section with identical teaching means information
Total duration obtains the corresponding students ' behavior of each teaching means information and analyzes the second parameter；
The second statistics subelement, is also used to according to the content of courses structure in the second type of teaching information, and
The corresponding temporal information of the target second audio section, counts the target second audio section with identical content of courses structure
Total duration obtains the corresponding students ' behavior analysis third parameter of each content of courses structure；
The second statistics subelement, is also used to according to the teaching means information in the second type of teaching information, system
The quantity with the second segment of target of identical teaching means information is counted, corresponding of each teaching means information is obtained
Raw the 4th parameter of behavioural analysis；
The second statistics subelement, is also used to according to the content of courses structure in the second type of teaching information, system
The quantity with the second segment of target of identical content of courses structure is counted, corresponding of each content of courses structure is obtained
Raw the 5th parameter of behavioural analysis；
Second generates subelement, for analyzing the first parameter, students ' behavior analysis second according to the students ' behavior
Parameter, students ' behavior analysis third parameter, the students ' behavior analyze the 4th parameter, students ' behavior analysis the 5th
Parameter generates the students ' behavior analysis result.
The third aspect of the embodiment of the present invention provides a kind of electronic equipment, including：Processor and memory, the processing
Device is connected with memory, wherein the memory supports electronic equipment to execute in first aspect of the embodiment of the present invention for storing
The program code of method, the processor are configured for executing the method in first aspect of the embodiment of the present invention.
Fourth aspect of the embodiment of the present invention provides a kind of computer storage medium, which is characterized in that the computer is deposited
Storage media is stored with computer program, and the computer program includes program instruction, refers to when the processor executes described program
The method in first aspect of the embodiment of the present invention is executed when enabling.
The first sound that the embodiment of the present invention is collected by obtaining the different radio reception device in two positions in classroom environment
Frequency information and the second audio-frequency information；The audio section for belonging to teacher's type is searched in the first audio-frequency information as the first sound of target
Frequency range is searched with the nonoverlapping audio section of temporal information of the first audio section of target in the second audio-frequency information as target the
Two audio sections；The first audio section of target is converted into the first text information of target, according to the identification of the first text information of target and mesh
Mark the corresponding first type of teaching information of the first audio section；According to the first type of teaching information, the time of the first audio section of target
Information, the audio types of target second audio section and audio frequency characteristics count the key link information in classroom environment.Due to
During handling audio-frequency information to obtain classroom key link information, mesh can be automatically searched without manually participating in
Mark with phonetic symbols frequency range and the type of teaching information for identifying target sound frequency range, then the key link information in programming count classroom environment, into
And it can be to avoid the bring tedious steps due to key link information of manual analysis classroom, to improve the analysis of Classroom instruction quality
Efficiency.Meanwhile using the alternate analysis between the first audio-frequency information and the second audio-frequency information, can rationally, accurate analysis class
Hall quality.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will to embodiment or
Attached drawing needed to be used in the description of the prior art is briefly described, it should be apparent that, the accompanying drawings in the following description is only
Some embodiments of the present invention, for those of ordinary skill in the art, without creative efforts, also
Other drawings may be obtained according to these drawings without any creative labor.
Fig. 1 a is a kind of flow diagram of multimedia data processing method provided in an embodiment of the present invention；
Fig. 1 b is a kind of schematic diagram of determining the first audio section of target provided in an embodiment of the present invention；
Fig. 2 is a kind of flow diagram for obtaining audio-frequency information provided in an embodiment of the present invention；
Fig. 3 is a kind of flow diagram for identifying type of teaching information provided in an embodiment of the present invention；
Fig. 4 is the flow diagram of another multimedia data processing method provided in an embodiment of the present invention；
Fig. 5 is the flow diagram of another multimedia data processing method provided in an embodiment of the present invention；
Fig. 6 is a kind of structural schematic diagram of apparatus for processing multimedia data provided in an embodiment of the present invention；
Fig. 7 is the structural schematic diagram of a kind of electronic equipment provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts it is all its
His embodiment, shall fall within the protection scope of the present invention.
Term " includes " in description and claims of this specification and above-mentioned attached drawing and " having " and they appoint
What is deformed, it is intended that is covered and non-exclusive is included.Such as contain the process, method of a series of steps or units, system,
Product or equipment are not limited to listed step or unit, but optionally further comprising the step of not listing or unit,
Or optionally further comprising other step or units intrinsic for these process, methods, product or equipment.
The execution of multimedia data processing method as mentioned in the embodiments of the present invention depends on computer program, can run
On the computer system of Feng Ruoyiman system.The computer program can integrate in the application, also can be used as independent work
Has class application operation.It is described in detail separately below：
A referring to Figure 1 is a kind of flow diagram of multimedia data processing method provided in an embodiment of the present invention, such as
Shown in Fig. 1 a, the multimedia data processing method is included at least：
Step S101 obtains the first audio-frequency information and the second audio-frequency information belonged in classroom environment；First audio
Information and second audio-frequency information are that the radio reception device for being located at different location by two respectively collects；
Specifically, the first audio-frequency information and the second audio-frequency information in classroom environment are extracted in audio database, wherein
First audio-frequency information is the audio-frequency information of the teacher collected by shotgun microphone, and shotgun microphone can be clearly collected into always
Sound of the teacher in classroom environment；Second audio-frequency information is the audio letter completely collected in classroom environment by omnidirectional microphone
Breath, shotgun microphone and omnidirectional microphone are placed on position different in classroom environment, for example, shotgun microphone can be placed
For collecting the audio-frequency information of teacher on dais before classroom, omnidirectional microphone can be placed on the wall among classroom and use
All audio-frequency informations in collection classroom environment.First audio-frequency information and the second audio-frequency information are with two-channel pulse code tune
The format storage for making (Pulse Code Modulation, PCM) stream, by specific data sample format with LLRR (Left
Left Right Right, left and right are right) it arranges to isolate the first audio-frequency information and the second audio-frequency information.First audio letter
Breath may include the audio-frequency information of teacher, the audio-frequency information of student, noise audio-frequency information and mute audio in classroom environment
Information；Second audio-frequency information may include the audio-frequency information of teacher (than the sound of the audio-frequency information of the teacher in the first audio-frequency information
Amount wants small), audio-frequency information (volume than the audio-frequency information of the student in the first audio-frequency information is big), the classroom environment of student
In noise audio-frequency information and mute audio information, the audio-frequency information of the teacher in certain first audio-frequency information it is the most clear；The
The audio-frequency information of student in two audio-frequency informations is the most clear.Since the first audio-frequency information and the second audio-frequency information are same
It is collected in a classroom environment in face of different objects, therefore the duration of the first audio-frequency information and the second audio-frequency information, and
Each moment, corresponding audio types (for example, teacher's audio types, student audio type) were all consistent in audio-frequency information,
In other words, it is complementary to one another between the content of the first audio-frequency information and the content of the second audio-frequency information.
Step S102 searches the audio section for belonging to teacher's type, as the first sound of target in first audio-frequency information
Frequency range, and target second audio section is searched in second audio-frequency information；The target second audio section and the target the
The corresponding temporal information of one audio section does not overlap；
Specifically, may include a variety of audio types in the first audio-frequency information (for example, teacher's type, single student's class
Type, mute type etc.) audio-frequency information search the first audio-frequency information therefore first with the classification feature of audio classification model
In belong to the audio section of teacher's type, as the first audio section of target, the audio section of teacher's type refers to that teacher's voice messaging accounts for
The audio section of major part, for example, the audio section that teacher explains trigonometric function is target first in the first audio-frequency information
Audio section.The nonoverlapping audio section of temporal information with the first audio section can also be further searched in the second audio-frequency information
As target second audio section, temporal information refers to the initial time stamp of audio section and terminates timestamp.It can be seen that target
The audio-frequency information of teacher's type is only included in first audio section, includes the audio of a variety of audio types in target second audio section
Information (for example, single student's type, more people student's types, mute type etc.), can be by the first audio by step S102
Information and the second audio-frequency information are divided into multiple the first audio sections of target and multiple target second audio sections.
The detailed process for searching the first audio section of target and target second audio section is：It can will be deposited in first audio section
In continuous sound and mute audio section as the first audio section of unit, be in the first audio section of unit otherwise be have connect
Continue the audio section of continuous sound or is mute audio section.Based on audio classification model, each the first sound of unit is identified respectively
The audio types of frequency range, audio types may include：Teacher's type, single student's type, more people student's types, mute type
Deng.It will identify that the first audio section of unit for belonging to teacher's type come is the first sound of target as the first audio section of target
The corresponding audio types of frequency range are teacher's type.It similarly, can be by there are continuous sound and mute sound in the second audio section
Frequency range is in the second audio section of unit or is wanted with the audio section of continuous continuous sound as the second audio section of unit
It is mute audio section.Based on above-mentioned audio classification model, the audio types of each the second audio of unit are identified respectively.It will know
Not Chu Lai the second audio section of unit for being not belonging to teacher's audio types, as audio section to be adjusted.If audio section to be adjusted
The temporal information of the first audio section of temporal information and target is not overlapped, then the audio section to be adjusted is just determined as target
Two audio sections；If the temporal information of the first audio section of temporal information and target of audio section to be adjusted has overlapping, then just will
The not corresponding audio section of lap in the audio section to be adjusted, is determined as target second audio section.
By taking first audio section of unit as an example, illustrate how audio types to be identified, to it based on audio classification model
Remaining the second audio section of the first audio section of unit or unit, using the available corresponding audio types of identical method.With
20ms is that window is long, and the intersection between window and window is 1/2, utilizes Fourier transformation from time domain number the first audio section of unit
According to being transformed to frequency domain data, low frequency enhanced processing is carried out by logarithmic function and is converted to Meier frequency spectrum, while by the frequency after conversion
Numeric field data is mapped to 64 dimensions spatially, then the first audio section of unit is converted to 64 × 501 matrix.The matrix of generation is defeated
Enter into trained audio classification model, the audio to match with the first audio section of unit is obtained from audio classification model
Type.It is understood that audio classification model be it is trained in advance, training process is an audio section and corresponding
Audio types are a training sample, pass through multiple audio sections and corresponding audio types (a plurality of training sample) building one
A audio classification model.It, can be from audio point after inputting the first audio section of unit to audio classification model after the completion of building
Matched audio types are exported in class model.For example, the first audio section of unit is：" for this problem, I thinks that A option is
Correctly ", by above-mentioned the first audio section of unit input audio disaggregated model, the output of audio classification model and above-mentioned unit the
One audio section " for this problem, I thinks that A option is correct " matched audio types are：Single student's type.
It is a kind of schematic diagram of determining the first audio section of target provided in an embodiment of the present invention please also refer to Fig. 1 b.Such as
Shown in Fig. 1 b, by when a length of 2 points of 30 seconds first audio-frequency information 10a be divided into 5 the first audio section of unit (5 units first
The temporal information of audio is respectively：00:00-00:20,00:20-01:10, 01:10-01:35,01:35-02:05,02:05-
02:30), by when a length of 2 points of 30 seconds second audio-frequency information 10b be divided into 4 the first audio sections of unit (4 the first sounds of unit
The temporal information of frequency is respectively： 00:00-00:45,00:45-01:35,01:35-02:05,02:05-02:30).It identifies respectively
The audio types of the first audio section of above-mentioned 5 units, wherein section 00:00-00:20 the first audio section of unit, section 01:
10-01:35 the first audio section of unit, section 02:05-02:The audio types of 30 the first audio section of unit are teacher's class
Type, therefore the first audio section of above-mentioned 3 units is the first audio section of target.The second audio section of above-mentioned 4 units is identified respectively
Audio types, wherein section 00:00-00:The audio types of 45 the second audio section of unit are single student's type, section
00:45-01:The audio types of 35 the second audio section of unit are more people student's types, section 01:35-02:05 unit
The audio types of two audio sections are mute type, section 02:05-02:The audio types of 30 the second audio section of unit are teacher
Type.To section 00:00-00:For 45 the second audio section of unit, due to section 00:00-00:20 audio section is target
First audio section, therefore, section 00:20-00:The audio types of 45 the second audio section of unit are single student's type.To area
Between 00:45-01:For 35 the second audio section of unit, due to section 01:10-01:35 audio section is the first audio of target
Section, therefore, section 00:45-01:The audio types of 10 the second audio section of unit are more people student's types.To section 01:35-
02:For 05 the second audio section of unit, section 01:35-02:The audio types of 05 the second audio section of unit are mute class
Type.To sum up, section 00:20-00:45 the second audio section of unit, section 00:45-01:10 the second audio section of unit, section
01:35-02:05 the second audio section of unit is target second audio section.
First audio section of target is converted to the first text information of target, according to the target first by step S103
Text information identifies type of teaching information corresponding with first audio section of target；
Specifically, identification the first audio section of target, obtains text information corresponding with the first audio section of target, referred to as mesh
Mark the first text information.Multiple the first audio sections of target are then respectively converted into text by multiple the first audio sections of target if it exists
After being converted to Textual information, text information continuous in time is pressed in order to improve the accuracy of type of teaching information for information
It is combined into the first text information of target according to timestamps ordering group, remaining text information is identified as the first text information of target.
The first text information of target is inputted in the first textual classification model, is obtained using the classification feature of the first textual classification model
Teaching means information corresponding with the first audio section of target；The first text information of target is inputted in the second textual classification model,
Content of courses structure corresponding with target second audio section is obtained using the classification feature of the second textual classification model, it will be above-mentioned
The teaching means information corresponding with the first audio section of target and content of courses structure got is used as the first type of teaching
Information.Teaching means information is that the strategy of education informations is mutually transmitted between teachers and students, and teaching means information may include：It reads, think
Examine, discuss, individual put question to or group put question to etc..Content of courses structure is the classroom link during teacher gives lessons, the content of courses
Structure may include：Evaluate feedback, studying new knowledge, connection consolidation, summary etc..For example, the first audio section of target is：" ask classmate
Read Third Nature section in textbook page 5, and think deeply two problems on blackboard ", first by above-mentioned the first audio of target
Section is converted to the first text information of target, by corresponding first text point of the first text information of target input teaching means information
In class model, identification " asks classmates to read textbook the 5th with the first audio-frequency information of target from above-mentioned first textual classification model
The corresponding teaching means information of Third Nature section in page, and think deeply two problems on blackboard " is：It reads；By target first
Text information inputs in corresponding second textual classification model of content of courses structure, knows from above-mentioned second textual classification model
" does not ask classmates to read the Third Nature section in textbook page 5 with the first audio-frequency information of target, and think deeply two on blackboard
The corresponding content of courses structure of problem " is：Studying new knowledge.
Optionally, speech recognition audio type belongs to the target second audio section of single student or more people student's types, obtains
To text information corresponding with target second audio section, referred to as the second text information of target.It will be with target the second text envelope manner of breathing
Adjacent the first text information of target and second text information of target is input to together in above-mentioned first textual classification model, from
It is obtained and the matched teaching means information of target second audio section in first textual classification model；It will be with the second text envelope of target
The first text information of target of manner of breathing neighbour and second text information of target are input to above-mentioned second textual classification model together
In, it is obtained and the matched content of courses structure of target second audio section from the second textual classification model.It is got above-mentioned
Teaching means information corresponding with target second audio section and content of courses structure are used as the first type of teaching information.Classroom
In environment, general teacher is in leading position, the target second audio section for belonging to single student or more people student's types it
Before, or later, certainly exist the audio-frequency information of teacher.Therefore, the first text of target and the second text of target are provided commonly for
The the second type of teaching information for analyzing target second audio section has reasonability, and the result accuracy rate analyzed is higher.
Step S104, according to the first type of teaching information, the corresponding temporal information of the first audio section of the target,
The corresponding temporal information of the target second audio section, the corresponding audio types of the target second audio section and audio frequency characteristics,
Count the key link information in the classroom environment；The key link information includes teacher's behavioural analysis result and student
Behavioural analysis result.
Specifically, according to teaching means information and content of courses structure, the first sound of target in the first type of teaching information
Teacher's behavioural analysis result in key link information in the corresponding temporal information statistics classroom environment of frequency range；According to second
Type of teaching information, the corresponding audio types of target second audio section and temporal information and audio frequency characteristics statistics key link letter
Students ' behavior analysis in breath is as a result, above-mentioned teacher's behavioural analysis result and students ' behavior analysis result are used to evaluate classroom matter
Amount.Audio frequency characteristics refer to the affective style for including according to the tone and decibel of audio identification audio, such as the audio pair of high tone
The audio frequency characteristics answered are high characteristic type, and the corresponding audio frequency characteristics of the audio of low decibel are low characteristic type.Teacher's behavior
Analysis is the result is that all teaching behaviors of the analysis teacher in classroom are resulting as a result, may include：Teacher's hours of instruction mention
Ask time to be answered, teacher's emotional arousal number etc.；Students ' behavior analysis is the result is that teaching of the analysis student in classroom participates in
Behavior is resulting as a result, may include：Student group answers number, student's think time, student classroom emotional arousal number
Deng.For example, statistics the first audio section of target sound intermediate frequency feature is the duration of high type feature, when as teacher's emotional arousal
Long, above-mentioned teacher's emotional arousal duration belongs to teacher's behavioural analysis result.
Further, Fig. 2 is referred to, is a kind of flow diagram for obtaining audio-frequency information provided in an embodiment of the present invention.
As shown in Fig. 2, the step of step S201- step S204, is retouched to the specific of step S101 in embodiment corresponding to above-mentioned Fig. 1 a
It states, i.e. the step of step S201- step S204 is a kind of detailed process for obtaining audio-frequency information provided in an embodiment of the present invention,
It can specifically include following steps：
Original first audio-frequency information cutting is original first audio-frequency information of multiple units by step S201, and by original the
Two audio-frequency information cuttings are original second audio-frequency information of multiple units；
Specifically, a length of 30ms when being multiple by the original first audio-frequency information cutting got from shotgun microphone
Original first audio-frequency information of unit, wherein original first audio-frequency information of each unit has 480 sampled points, unit original
The sample rate of one audio-frequency information is 16000Hz.Similarly, the original second audio-frequency information cutting that will be got from omnidirectional microphone
The unit of a length of 30ms original second audio-frequency information when being multiple, original second audio-frequency information of each unit have 480 samplings
Point, the sample rate of original second audio-frequency information of unit are 16000Hz.
Step S202 calculates the noise probability coefficent of the subband in original first audio-frequency information of the unit；
Specifically, first including the original first audio-frequency information frequency reducing of unit of 480 sampled points, 16000Hz by each
For original first audio-frequency information of unit of 240 sampled points, 8000Hz.80Hz- in original first audio-frequency information of unit of account
The noise probability of 250Hz, 250Hz-500Hz, 500Hz-1000Hz, 1000Hz-2000Hz, 2000Hz-3000Hz6 subbands
Coefficient, calculation method be mainly pass through Gauss Markov model calculate each subband noise probability coefficent and voice it is general
Rate coefficient.The speech probability P of each subband is calculated first with formula (1.1)speech(x) and formula (1.2) calculating is each
The noise probability P of a subbandnoisy(x),
Wherein, σspeechFor the corresponding voice variance of subband,For the corresponding speech mean of subband；σnoisyFor subband
Corresponding noise variance,For the corresponding noise mean value of subband.Voice variances sigmaspeech, speech meanNoise variance
σnoisy, noise mean valueIt is by initializing in advance.I is the integer between 1 to 6, indicates the corresponding language of each subband
Sound probability and noise probability.
The speech probability P of each subband is calculatedspeech(x) and noise probability Pnoisy(x) after, formula is utilized
(1.3) in original first audio-frequency information of unit of account each subband noise probability coefficent Ls(xi)：
Ls(xi) indicate original first audio-frequency information of unit in each subband noise probability coefficent.
Step S203, if the noise probability coefficent is greater than probability threshold value, by original first audio-frequency information of the unit
It deletes, and unit original second audio-frequency information Chong Die with the temporal information of original first audio-frequency information of the unit is deleted；
Specifically, as long as there are the noise probability coefficent L of a subband in original first audio-frequency information of units(xi) be greater than
Probability threshold value, as long as that is to say that there are the noise probability coefficents of a subband to be greater than generally in the noise probability coefficent of 6 subbands
Rate threshold value determines that the noise probability coefficent of corresponding original first audio-frequency information of unit is greater than probability threshold value (corresponding unit
Original first audio-frequency information is noise types).And original first audio-frequency information of unit that will be greater than probability threshold value is from original first
It is deleted in audio-frequency information, teacher's audio-frequency information, student audio information and mute audio is only retained in original first audio-frequency information
Information is conducive to subsequent voice and is converted to the accuracy rate of text and obtains type of teaching letter from convolutional neural networks model
The accuracy of breath.In order to make the duration of original second audio-frequency information and delete original the of original first audio-frequency information of unit
The duration of one audio-frequency information is equal, only need to will be Chong Die with the temporal information of original first audio-frequency information of the unit of noise types
Original second audio-frequency information of unit is deleted, without the noise probability coefficent of subband in original second audio-frequency information of unit of account.
For example, original first audio-frequency information can the original first audio-frequency information A as unit of cutting, the original first audio-frequency information B of unit,
The original first audio-frequency information C of unit, if the noise probability coefficent of the original first audio-frequency information B of unit is greater than threshold value, by unit
Original first audio-frequency information B is deleted from original first audio-frequency information；It, will be with unit meanwhile in original second audio-frequency information
The original second audio-frequency information D of unit of the temporal information overlapping of original first audio-frequency information B is deleted.
Step S204 merges original first audio-frequency information of remaining unit according to the sequence of timestamp, as institute
The first audio-frequency information is stated, and original second audio-frequency information of remaining unit is merged according to the sequence of timestamp, as institute
State the second audio-frequency information.
Specifically, after deleting noise probability coefficent greater than original first audio-frequency information of unit of probability threshold value, it will be remaining
Original first audio-frequency information of unit merged according to the timestamps ordering in audio-frequency information, as the first audio-frequency information；Together
Reason, original second audio-frequency information of remaining unit is merged according to the timestamps ordering in audio-frequency information, as the second sound
Frequency information.By deleting original first audio-frequency information of unit of noise types, in the first audio-frequency information and the second audio-frequency information only
Retain non-noise audio information portion (for example, teacher's audio-frequency information, student audio information, mute audio information etc.).
Further, Fig. 3 is referred to, is that a kind of process for identifying type of teaching information provided in an embodiment of the present invention is shown
It is intended to.As shown in figure 3, the step of step S301- step S303 is the tool to step S103 in embodiment corresponding to above-mentioned Fig. 1 a
Body description, i.e. the step of step S301- step S303 are a kind of tools for identifying type of teaching information provided in an embodiment of the present invention
Body process, may include steps of：
First audio section of target is converted to the first text information of target by step S301, by first text of target
The first textual classification model of this information input obtains and the matched teaching means information of the first audio section of target；
Specifically, speech recognition first audio section of target, obtains the first text information of target.Be converted to target first
After text information, the first text information of target after conversion is inputted into the first textual classification model, first textual classification model
It can be convolutional neural networks model, the teaching to match with the first audio section of target obtained from the first textual classification model
Means Information.It is understood that above-mentioned first textual classification model complete in advance by building, building process can be one
Textual information and a teaching means information are a training data, and the normal distribution in [- 1,1] initializes term vector at random
Matrix, above-mentioned term vector matrix are 128 dimensions；The trained text that is used for of input is initialized, convolutional layer is then passed to
Convolution is carried out, convolutional layer is 3 layers, and each layer of convolution kernel is respectively：3 × 3,5 × 5,8 × 8；Pond layer is transferred to after convolution,
The order of magnitude of feature after convolution is reduced using maximum pondization；The output of last each kernel is connected to a vector, and transmits
To full articulamentum, a text information the first textual classification model corresponding with teaching means information is constructed.First text classification
After model foundation, text information is inputted into the first textual classification model, so that it may get corresponding teaching means information.
First text information of target is inputted the second textual classification model by step S302, is obtained and the target the
The matched content of courses structure of one audio section；
Specifically, after the first audio section of target is converted to the first text information of target, by the first text of target after conversion
In the second textual classification model of information input, which is also possible to convolutional neural networks model, from second
The content of courses structure to match with the first audio section of target is obtained in textual classification model.Second textual classification model has mentioned
Preceding building is completed, and building process may refer to the process that the first textual classification model is constructed in the step S301 in above-mentioned Fig. 3.
After second textual classification model is established, text information is inputted into the second textual classification model, so that it may get corresponding religion
Learn content structure.Teaching means information is obtained according to the first text information of target and content of courses structure is independent from each other two
A step, that is to say, that before the step S301 in Fig. 3 can execute the step S302 in Fig. 3, or execute in step
After S302 or step S301 and step S302 is performed simultaneously.
Step S303, by the corresponding teaching means information of the first audio section of target and the first audio section of the target
Corresponding content of courses structure determination is the first type of teaching information corresponding with first audio section of target.
Specifically, the teaching means information that will be obtained from the first textual classification model, and from the second textual classification model
The content of courses structure determination of middle acquisition is the corresponding type of teaching information of the first audio section of target.If there is multiple targets first
Audio section executes above-mentioned steps to each the first audio section of target respectively, obtains the corresponding teaching of each the first audio section of target
Type information.
The first audio-frequency information and second collected by obtaining the different radio reception device in two positions in classroom environment
Audio-frequency information；The audio section for belonging to teacher's type is searched in the first audio-frequency information as the first audio section of target, in the second sound
The nonoverlapping audio section of temporal information with the first audio section of target is searched in frequency information as target second audio section；By mesh
It marks the first audio section and is converted to the first text information of target, according to the identification of the first text information of target and the first audio section of target
Corresponding first type of teaching information；According to the first type of teaching information, the temporal information of the first audio section of target, target
The audio types type and audio frequency characteristics of two audio sections count the key link information in classroom environment.Due to believing to audio
During breath processing is to obtain classroom key link information, target sound frequency range can be automatically searched without manually participating in
And identify the type of teaching information of target sound frequency range, then the key link information in programming count classroom environment, and then can keep away
Exempt from the bring tedious steps due to key link information of manual analysis classroom, to improve the efficiency of analysis Classroom instruction quality.Together
When, using the alternate analysis between the first audio-frequency information and the second audio-frequency information, can rationally, accurately analyze Classroom instruction quality.
Further, Fig. 4 is referred to, is the stream of another multimedia data processing method provided in an embodiment of the present invention
Journey schematic diagram.As shown in figure 4, multimedia data processing method may include：
Step S401 obtains the first audio-frequency information and the second audio-frequency information belonged in classroom environment；
Step S402 searches the audio section for belonging to teacher's type, as the first sound of target in first audio-frequency information
Frequency range, and target second audio section is searched in second audio-frequency information；The target second audio section and the target the
The corresponding temporal information of one audio section does not overlap；
First audio section of target is converted to the first text information of target, according to the target first by step S403
Text information identifies the first type of teaching information corresponding with first audio section of target；
Step S404, according to the first type of teaching information, the corresponding temporal information of the first audio section of the target,
The corresponding temporal information of the target second audio section, the corresponding audio types of the target second audio section and audio frequency characteristics,
Count the key link information in the classroom environment；The key link information includes teacher's behavioural analysis result and student
Behavioural analysis result；
Wherein, the detailed process of step S401- step S404 may refer to the step S101- step in above-mentioned Fig. 1 a
The description of S104, and the detailed process for obtaining audio-frequency information may refer to retouching for the step S201- step S204 in above-mentioned Fig. 2
It states, the detailed process of the first type of teaching information of identification may refer to the description of the step S301- step S303 in above-mentioned Fig. 3,
It is no longer repeated herein.
Step S405 obtains the index model to match with the classroom environment, and the index model is by teaching notes structure
The knowledge-point models in classroom built；
Specifically, obtain in advance building complete with the matched index model of the first audio-frequency information, index model is to pass through
Knowledge-point models in the classroom of teaching notes building, therefore include knowledge point in index model.For example, with the first audio-frequency information
The index model matched is the index model about single vowel, therefore knowledge point included in the index model about single vowel
For：Tone, a sound, two sound, three sound, the four tones of standard Chinese pronunciation.
Step S406 extracts the knowledge point text in the index model, in first text information of target, really
The fixed object time with the text informations of the knowledge point text matches stabs, and when target in first audio-frequency information
Between stab position addition Keyword Tag；Text in the Keyword Tag is identical as the knowledge point text；
Specifically, the knowledge point text in index model is successively extracted, by knowledge point text and the first text information of target
In word carry out the matching based on text string editing distance, if in the first text information of target exist and knowledge point text
The word matched, then extract the timestamp of matched word stabbed as the object time, above-mentioned target in the first audio-frequency information
Keyword Tag is added in the position of timestamp, and the text in the Keyword Tag is identical as knowledge point text, is known with establishing one
Point is known to the index between teacher's audio-frequency information.Matched detailed process is：In calculation knowledge point text and target text information
The editing distance of word regard the shorter text of text size in knowledge point text and above-mentioned word as benchmark text.Judgement
Whether the ratio of editing distance and the text size of benchmark text is less than preset fractional threshold (in general, fractional threshold is small
In 25%), if being less than fractional threshold, it is determined that the word is matched with knowledge point text.For example, knowledge point text
The editing distance of A and the word B in target text information are 1, and the text size of knowledge point text A is 5, and the text of word B is long
Degree is 6, therefore benchmark text is knowledge point text A, text size 5, due to the text of editing distance and knowledge point text A
The ratio of length is less than 25% (1/5=20% of fractional threshold<25%), therefore knowledge point text A and word B is matched.
Step S407, when getting searching keyword, retrieval and the inquiry are crucial in first audio-frequency information
The matched Keyword Tag of word, as target keywords label；
Specifically, user can work as acquisition according to the knowledge point in index model from input module input inquiry keyword
To user input searching keyword when, in the first audio-frequency information retrieval with the matched keyword mark of above-mentioned key word of the inquiry
Label, as target keywords label.Retrieving equally can be to close in the searching keyword and Keyword Tag that calculate input
Editing distance between key word can determine the keyword and keyword mark when editing distance is less than preset distance threshold
Label are matched.If there is no the Keyword Tags with Keywords matching in the first audio-frequency information, sends and be not present to user
The prompting message of corresponding audio-frequency information.
Step S408 is exported corresponding with the timestamp of the target keywords label in first audio-frequency information
Audio-frequency information, and determine the curriculum information of audio-frequency information corresponding with the target keywords label.
Specifically, audio-frequency information corresponding with the timestamp of target keywords label is searched in the first audio-frequency information,
And above-mentioned corresponding audio-frequency information is exported, while the course letter of audio-frequency information corresponding with target keywords label can be exported
Breath, for example, name, academic title and the course hours of output teacher.If target keywords label more than one, according to
The height of the matching degree of target keywords label and the searching keyword of input exports corresponding audio-frequency information, that is to say, that
The matching degree of target keywords label and the searching keyword of input is higher, then what is exported is corresponding with target keywords label
Audio-frequency information sequence it is more forward.
The first audio-frequency information and second collected by obtaining the different radio reception device in two positions in classroom environment
Audio-frequency information；The audio section for belonging to teacher's type is searched in the first audio-frequency information as the first audio section of target, in the second sound
The nonoverlapping audio section of temporal information with the first audio section of target is searched in frequency information as target second audio section；By mesh
It marks the first audio section and is converted to the first text information of target, according to the identification of the first text information of target and the first audio section of target
Corresponding first type of teaching information；According to the first type of teaching information, the temporal information of the first audio section of target, target
The audio types and audio frequency characteristics of two audio sections count the key link information in classroom environment.Due to audio-frequency information
During reason is to obtain classroom key link information, without manually participating in automatically searching target sound frequency range and know
The type of teaching information of other target sound frequency range, then the key link information in programming count classroom environment, so can to avoid because
Manual analysis classroom key link information and bring tedious steps, to improve the efficiency of analysis Classroom instruction quality.Meanwhile it adopting
With the alternate analysis between the first audio-frequency information and the second audio-frequency information, can rationally, accurately analyze Classroom instruction quality, and build
Knowledge point in vertical teaching notes can quickly position the Keywords matching with input to the index relative between the first audio-frequency information
Audio-frequency information, improve classroom environment in audio recall precision.
Further, Fig. 5 is referred to, is the stream of another multimedia data processing method provided in an embodiment of the present invention
Journey schematic diagram.As shown in figure 5, multimedia data processing method may include steps of：
Step S501 obtains the first audio-frequency information and the second audio-frequency information belonged in classroom environment；
Step S502 searches the audio section for belonging to teacher's type, as the first sound of target in first audio-frequency information
Frequency range, and target second audio section is searched in second audio-frequency information；
First audio section of target is converted to the first text information of target, according to the target first by step S503
Text information identifies the first type of teaching information corresponding with first audio section of target；
Wherein, the detailed process of step S501- step S503 may refer to the step S101- step in above-mentioned Fig. 1 a
The description of S103, and the detailed process for obtaining audio-frequency information may refer to retouching for the step S201- step S204 in above-mentioned Fig. 2
It states, the detailed process of the first type of teaching information of identification may refer to the description of the step S301- step S303 in above-mentioned Fig. 3,
It is no longer repeated herein.
Step S504, according to the first type of teaching information, the corresponding temporal information of the first audio section of the target,
Generate teacher's behavioural analysis result；
Specifically, statistics in the first type of teaching information, has the first audio section of target of identical teaching means information
Total duration, as corresponding the first parameter of teacher's behavioural analysis of the teaching means information (for example, statistics " seminar " religion
The total duration for learning to do segment information, as the first parameter of teacher's behavioural analysis)；Statistics has phase in the first type of teaching information
With the total duration of the first audio section of target of content of courses structure, as the corresponding teacher's behavioural analysis of the content of courses structure
Second parameter (for example, total duration of statistics " studying new knowledge " content of courses structure, as the second parameter of teacher's behavioural analysis)；
Statistics is in the first type of teaching information, the quantity of the first audio section of target with identical teaching means information, as the religion
The corresponding teacher's behavioural analysis third parameter of segment information is learned to do (for example, the target of statistics " seminar " teaching means information
The quantity of first audio, as teacher's behavioural analysis third parameter)；Meter has identical teaching in the first type of teaching information
The quantity of the first audio section of target of content structure, as corresponding the 4th parameter of teacher's behavioural analysis of the content of courses structure
(for example, the quantity of first audio of target of statistics " studying new knowledge " content of courses structure, joins as teacher's behavioural analysis the 4th
Number).According to above-mentioned the first parameter of teacher's behavioural analysis, the second parameter of teacher's behavioural analysis, teacher's behavioural analysis third parameter,
The 4th parameter of teacher's behavioural analysis generates teacher's behavioural analysis as a result, teacher's behavioural analysis result may include：
It is that give lessons duration, classroom of teacher reviews one's lessons duration, seminar duration, puts question to duration to be answered and teacher's emotional arousal duration etc..
Step S505, according to the second type of teaching information, the corresponding temporal information of the target second audio section,
The corresponding audio types of the target second audio section and audio frequency characteristics generate students ' behavior and analyze result；
Specifically, statistics in the second type of teaching information, has the target second of identical audio frequency characteristics and audio types
The quantity of audio section analyzes the first parameter (for example, statistics " single student " audio types, " being heightened in spirits " as students ' behavior
The quantity of the target second audio section of audio frequency characteristics analyzes the first parameter as students ' behavior)；Statistics is in the second type of teaching
In information, the total duration of the target second audio section with identical teaching means information is corresponding as the teaching means information
Students ' behavior analyzes the second parameter (for example, the total duration of statistics " individual is putd question to " teaching means information, divides as students ' behavior
Analyse the second parameter)；Statistics has the target second audio section of identical content structure information in the second type of teaching information
Total duration, as the corresponding students ' behavior analysis third parameter of the content of courses structure (for example, statistics " classroom summary " teaching
The total duration of content structure analyzes third parameter as students ' behavior)；Statistics has identical in the second type of teaching information
The quantity of the target second audio section of teaching means information, as the corresponding students ' behavior analysis the 4th of the teaching means information
Parameter (for example, the audio segment number of statistics " individual is putd question to " teaching means information, analyzes the 4th parameter as students ' behavior)；
Statistics is in the second type of teaching information, the quantity of the target second audio section with identical content of courses structure, as the religion
It learns the corresponding students ' behavior of content structure and analyzes the 5th parameter (for example, the audio section of statistics " classroom summary " content of courses structure
Quantity analyzes the 5th parameter as students ' behavior).The first parameter, the students ' behavior point are analyzed according to above-mentioned students ' behavior
It analyses the second parameter, students ' behavior analysis third parameter, the students ' behavior and analyzes the 4th parameter, the students ' behavior point
The 5th parameter is analysed, generates the students ' behavior analysis as a result, students ' behavior analysis result may include：It is student's emotional arousal
Number, review one's lessons duration, seminar duration and student group answer number etc..
Teacher's behavioural analysis result and students ' behavior analysis result are determined as the classroom by step S506
Key link information in environment.
Specifically, the teacher's behavior outcome and students ' behavior result that above-mentioned steps generate are determined as in classroom environment
Key link information.
The first audio-frequency information and second collected by obtaining the different radio reception device in two positions in classroom environment
Audio-frequency information；The audio section for belonging to teacher's type is searched in the first audio-frequency information as the first audio section of target, in the second sound
The nonoverlapping audio section of temporal information with the first audio section of target is searched in frequency information as target second audio section；By mesh
It marks the first audio section and is converted to the first text information of target, according to the identification of the first text information of target and the first audio section of target
Corresponding first type of teaching information；According to the first type of teaching information, the temporal information of the first audio section of target, target
The audio types and audio frequency characteristics of two audio sections count the key link information in classroom environment.Due to audio-frequency information
During reason is to obtain classroom key link information, without manually participating in automatically searching target sound frequency range and know
The type of teaching information of other target sound frequency range, then the key link information in programming count classroom environment, so can to avoid because
Manual analysis classroom key link information and bring tedious steps, to improve the efficiency of analysis Classroom instruction quality.Meanwhile it adopting
With the alternate analysis between the first audio-frequency information and the second audio-frequency information, can rationally, accurately analyze Classroom instruction quality, and build
Knowledge point in vertical teaching notes can quickly position the Keywords matching with input to the index relative between the first audio-frequency information
Audio-frequency information, improve classroom environment in audio recall precision.
Further, Fig. 6 is referred to, is a kind of structure of apparatus for processing multimedia data provided in an embodiment of the present invention
Schematic diagram.As shown in fig. 6, the apparatus for processing multimedia data 1 includes at least：First obtains module 11, the first searching module
12, the second searching module 13, conversion module 14, identification module 15, statistical module 16；
First obtains module 11, for obtaining the first audio-frequency information and the second audio-frequency information that belong in classroom environment；Institute
It states the first audio-frequency information and second audio-frequency information is that the radio reception device for being located at different location by two respectively collects；
First searching module 12, for searching the audio section for belonging to teacher's type in first audio-frequency information, as
The first audio section of target；
Second searching module 13, for searching target second audio section in second audio-frequency information；The target
Two audio sections temporal information corresponding with first audio section of target does not overlap；
Conversion module 14, for first audio section of target to be converted to the first text information of target；
Identification module 15, for according to target text information identification corresponding with first audio section of target the
One type of teaching information；
Statistical module 16, for according to the first type of teaching information, the first audio section of the target corresponding time
Information, the corresponding temporal information of the target second audio section, the corresponding audio types of the target second audio section and audio
Feature counts the key link information in the classroom environment；The key link information include teacher's behavioural analysis result and
Students ' behavior analyzes result.
Wherein, first module 11, the first searching module 12, the second searching module 13, conversion module 14, identification mould are obtained
Block 15, statistical module 16 specific implementation can be found in embodiment corresponding to above-mentioned Fig. 1 a to step S101- step
S104 is not discussed here.
Further, Fig. 6 is referred to, apparatus for processing multimedia data 1 can also include：Determining module 17；
The conversion module 14 is also used to target that audio types are single student's type or more people student's types the
Two audio sections are converted to the second text information of target, by target first text envelope adjacent with second text information of target
Breath and second text information of target are input in first textual classification model, are obtained and the target second audio
The matched teaching means information of section；
First obtains module 11, is also used to target first text information adjacent with second text information of target
It is input in second textual classification model, obtains and the target second audio section with second text information of target
Matched content of courses structure；
Determining module 17 is used for the corresponding teaching means information of the target second audio section and the target second
The corresponding content of courses structure determination of audio section is the second type of teaching information corresponding with the target second audio section.
Wherein, first acquisition module 11, conversion module 14, determining module 17 specific implementation can be found in above-mentioned figure
To step S103 in embodiment corresponding to 1a, it is not discussed here.
Further, Fig. 6 is referred to, apparatus for processing multimedia data 1 may include：First acquisition module 11, first are looked into
Look for module 12, the second searching module 13, conversion module 14, identification module 15, statistical module 16, determining module 17；It can also wrap
It includes：Second obtains module 18, adding module 19, retrieval module 20, output module 21.
Second obtains module 18, for obtaining the index model to match with the classroom environment；The index model is
The knowledge-point models in classroom constructed by teaching notes；
Adding module 19, for extracting the knowledge point text in the index model, in first text information of target
In, the determining object time with the text information of the knowledge point text matches stabs, and the mesh in first audio-frequency information
Add Keyword Tag in the position for marking timestamp；Text in the Keyword Tag is identical as the knowledge braille text；
Retrieval module 20, for retrieving in first audio-frequency information and being looked into described when getting searching keyword
The Keyword Tag for asking Keywords matching, as target keywords label；
Output module 21, for exporting the timestamp with the target keywords label in first audio-frequency information
Corresponding audio-frequency information, and determine the curriculum information of audio-frequency information corresponding with the target keywords label.
Wherein, the second acquisition module 18, adding module 19, retrieval module 20, the specific implementation of output module 21 can
Referring to, to step S405- step S408, being not discussed here in embodiment corresponding to above-mentioned Fig. 4.
As shown in fig. 6, the first searching module 12 may include：First cutting unit 121, the first recognition unit 122,
One determination unit 123；
First cutting unit 121 is used to according to audio frequency characteristics be multiple units first by the first audio-frequency information cutting
Audio section；Each the first audio section of unit is the audio section comprising continuous sound or is mute audio section；
First recognition unit 122, for each first audio section of unit to be input in audio classification model, point
It Shi Bie not be with the matched audio types of each first audio section of unit；
First determination unit 123, for being teacher's type by the audio types in first audio-frequency information
The first audio section of unit, be determined as first audio section of target.
Wherein, the first cutting unit 121, the first recognition unit 122, the first determination unit 123 specific implementation can
Referring to, to step S102, being not discussed here in embodiment corresponding to above-mentioned Fig. 1 a.
As shown in fig. 6, the second searching module 13 may include：Second cutting unit 131, the second recognition unit 132,
Two determination units 133；
Second cutting unit 131 is used to according to audio frequency characteristics be multiple the second audios of unit by the second audio-frequency information cutting
Section；Each the second audio section of unit is the audio section comprising continuous sound or is mute audio section；
Second recognition unit 132, for each second audio section of unit to be input to the audio classification model
In, it identifies and the matched audio types of each second audio section of unit respectively；
Second determination unit 133, for not being teacher's type by the audio types in second audio-frequency information
The second audio section of unit, as audio section to be adjusted；
Second determination unit 133, if being also used to the temporal information and first audio of target of audio section to be adjusted
The temporal information of section is not overlapped, it is determined that if being target second audio section for the audio section to be adjusted；
Second determination unit 133, if being also used to the temporal information and first audio of target of audio section to be adjusted
The temporal information overlapping of section is determined as the target then by the not corresponding audio section of lap in the audio section to be adjusted
Second audio section.
Wherein, the second cutting unit 131, the second recognition unit 132, the second determination unit 133 specific implementation can
Referring to, to step S102, being not discussed here in embodiment corresponding to above-mentioned Fig. 1 a.
As shown in fig. 6, identification module 15 may include：Acquiring unit 151, third determination unit 152；
Acquiring unit 151, for first text information of target to be inputted the first textual classification model, acquisition and institute
State the matched teaching means information of the first audio section of target；
The acquiring unit 151 is obtained for first text information of target to be inputted the second textual classification model
With the matched content of courses structure of the first audio section of target；
Third determination unit 152 is used for the corresponding teaching means information of the first audio section of target and the target
The corresponding content of courses structure determination of first audio section is the first type of teaching corresponding with first audio section of target letter
Breath.
Wherein, acquiring unit 151, the specific implementation of third determination unit 152 can be found in real corresponding to above-mentioned Fig. 3
It applies in example to step S301- step S303, is not discussed here.
As shown in fig. 6, statistical module 16 may include：First generation unit 161, the second generation unit the 162, the 4th are really
Order member 163；
First generation unit 161, for corresponding according to the first type of teaching information, the first audio section of the target
Temporal information, generate teacher's behavioural analysis result；
Second generation unit 162, for corresponding according to the second type of teaching information, the target second audio section
Temporal information, the corresponding audio types of the target second audio section and audio frequency characteristics, generate students ' behavior analyze result；
4th determination unit 163, for determining teacher's behavioural analysis result and students ' behavior analysis result
For the key link information in the classroom environment.
Wherein, the first generation unit 161, the second generation unit 162, the specific implementation of the 4th determination unit 163 can
Referring to, to step S504- step S506, being not discussed here in embodiment corresponding to above-mentioned Fig. 5.
As shown in fig. 6, the first generation unit 161 may include：First statistics subelement 1611, first generates subelement
1612；
First statistics subelement 1611, for according in the first type of teaching information teaching means information and institute
The corresponding temporal information of the first audio section of target is stated, the total of first audio section of target with identical teaching means information is counted
Duration obtains corresponding the first parameter of teacher's behavioural analysis of each teaching means information；
The first statistics subelement 1611, is also used to according to the teaching hand content in the first type of teaching information
Structure and the corresponding temporal information of the first audio section of the target count first sound of target with identical content of courses structure
The total duration of frequency range obtains corresponding the second parameter of teacher's behavioural analysis of each content of courses structure；
The first statistics subelement 1611 is also used to according to the teaching means letter in the first type of teaching information
Breath counts the quantity with the first segment of target of identical teaching means information, obtains each teaching means information and respectively correspond
Teacher's behavioural analysis third parameter；
The first statistics subelement 1611, is also used to according to the content of courses knot in the first type of teaching information
Structure counts the quantity with the first segment of target of identical content of courses structure, obtains each content of courses structure and respectively correspond
The 4th parameter of teacher's behavioural analysis；
First generates subelement 1612, for according to first parameter of teacher's behavioural analysis, teacher's behavioural analysis second
Parameter, teacher's behavioural analysis third parameter, the 4th parameter of teacher's behavioural analysis, generate teacher's behavioural analysis result.
Wherein, the specific implementation that the first statistics subelement 1611, first generates subelement 1612 can be found in above-mentioned figure
To step S504 in embodiment corresponding to 5, it is not discussed here.
As shown in fig. 6, the second generation unit 162 may include：Second statistics subelement 1621, second generates subelement
1622；
Second statistics subelement 1621, for counting the target second audio with identical audio frequency characteristics and audio types
The quantity of section analyzes the first parameter as students ' behavior；
The second statistics subelement 1621 is also used to according to the teaching means letter in the second type of teaching information
Temporal information corresponding with the target second audio section is ceased, the target second audio with identical teaching means information is counted
The total duration of section obtains the corresponding students ' behavior of each teaching means information and analyzes the second parameter；
The second statistics subelement 1621, is also used to according to the content of courses knot in the second type of teaching information
Structure and the corresponding temporal information of the target second audio section count the target second audio with identical content of courses structure
The total duration of section obtains the corresponding students ' behavior analysis third parameter of each content of courses structure；
The second statistics subelement 1621 is also used to according to the teaching means letter in the second type of teaching information
Breath counts the quantity with the second segment of target of identical teaching means information, obtains each teaching means information and respectively correspond
Students ' behavior analyze the 4th parameter；
The second statistics subelement 1621, is also used to according to the content of courses knot in the second type of teaching information
Structure counts the quantity with the second segment of target of identical content of courses structure, obtains each content of courses structure and respectively correspond
Students ' behavior analyze the 5th parameter；
Second generates subelement 1622, and for analyzing the first parameter according to the students ' behavior, the students ' behavior is analyzed
Second parameter, students ' behavior analysis third parameter, the students ' behavior analyze the 4th parameter, students ' behavior analysis
5th parameter generates the students ' behavior analysis result.
Wherein, the specific implementation that the second statistics subelement 1621, second generates subelement 1622 can be found in above-mentioned figure
To step S505 in embodiment corresponding to 5, it is not discussed here.
The first audio-frequency information and second collected by obtaining the different radio reception device in two positions in classroom environment
Audio-frequency information；The audio section for belonging to teacher's type is searched in the first audio-frequency information as the first audio section of target, in the second sound
The nonoverlapping audio section of temporal information with the first audio section of target is searched in frequency information as target second audio section；By mesh
It marks the first audio section and is converted to the first text information of target, according to the identification of the first text information of target and the first audio section of target
Corresponding first type of teaching information；According to the first type of teaching information, the temporal information of the first audio section of target, target
The audio types and audio frequency characteristics of two audio sections count the key link information in classroom environment.Due to audio-frequency information
During reason is to obtain classroom key link information, without manually participating in automatically searching target sound frequency range and know
The type of teaching information of other target sound frequency range, then the key link information in programming count classroom environment, so can to avoid because
Manual analysis classroom key link information and bring tedious steps, to improve the efficiency of analysis Classroom instruction quality.Meanwhile it adopting
With the alternate analysis between the first audio-frequency information and the second audio-frequency information, can rationally, accurately analyze Classroom instruction quality, and build
Knowledge point in vertical teaching notes can quickly position the Keywords matching with input to the index relative between the first audio-frequency information
Audio-frequency information, improve classroom environment in audio recall precision.
Further, Fig. 7 is referred to, is the structural schematic diagram of a kind of electronic equipment provided in an embodiment of the present invention.Such as figure
Shown in 7, the electronic equipment 1000 may include：Processor 1001, network interface 1004 and memory 1005, in addition, described
Electronic equipment 1000 can also include：User interface 1003 and at least one communication bus 1002.Wherein, communication bus 1002
For realizing the connection communication between these components.Wherein, user interface 1003 may include display screen (Display), keyboard
(Keyboard), optional user interface 1003 can also include standard wireline interface and wireless interface.Network interface 1004 can
Choosing may include standard wireline interface and wireless interface (such as WI-FI interface).Memory 1005 can be high-speed RAM and deposit
Reservoir is also possible to non-labile memory (non-volatile memory), for example, at least a magnetic disk storage.It deposits
Reservoir 1005 optionally can also be that at least one is located remotely from the storage device of aforementioned processor 1001.As shown in fig. 7, making
To may include operating system, network communication module, Subscriber Interface Module SIM in a kind of memory 1005 of computer storage medium
And equipment controls application program.
In electronic equipment 1000 shown in Fig. 7, network interface 1004 can provide network communication function；And user interface
1003 are mainly used for providing the interface of input for user；And processor 1001 can be used for calling and store in memory 1005
Equipment controls application program, to realize：
Obtain the first audio-frequency information and the second audio-frequency information belonged in classroom environment；First audio-frequency information and institute
Stating the second audio-frequency information is that the radio reception device for being located at different location by two respectively collects；
The audio section for belonging to teacher's type is searched in first audio-frequency information, as the first audio section of target, and
Target second audio section is searched in second audio-frequency information；The target second audio section and first audio section of target
Corresponding temporal information does not overlap；
First audio section of target is converted into the first text information of target, according to first text information of target
Identify the first type of teaching information corresponding with first audio section of target；
According to the first type of teaching information, the corresponding temporal information of the first audio section of the target, the target
The corresponding temporal information of two audio sections, the corresponding audio types of the target second audio section and audio frequency characteristics, count the class
Key link information in hall environment；The key link information includes teacher's behavioural analysis result and students ' behavior analysis knot
Fruit.
In one embodiment, the processor 1001 is searched in first audio-frequency information in execution belongs to teacher
The audio section of type specifically executes following steps when as the first audio section of target：
According to audio frequency characteristics by the first audio-frequency information cutting be multiple the first audio sections of unit；Each unit first
Audio section is the audio section comprising continuous sound or is mute audio section；
Each first audio section of unit is input in audio classification model, respectively identification and each unit
The matched audio types of first audio section；
It is the first audio section of unit of teacher's type by the audio types in first audio-frequency information,
It is determined as first audio section of target.
In one embodiment, the processor 1001 is executing the lookup target second in second audio-frequency information
When audio section, following steps are specifically executed：
According to audio frequency characteristics by the second audio-frequency information cutting be multiple the second audio sections of unit；Each the second audio of unit
Section is the audio section comprising continuous sound or is mute audio section；
Each second audio section of unit is input in the audio classification model, is identified respectively and described each
The matched audio types of the second audio section of unit；
It is not the second audio section of unit of teacher's type by the audio types in second audio-frequency information, as
Audio section to be adjusted；
If the temporal information of audio section to be adjusted and the temporal information of first audio section of target be not be overlapped, it is determined that
If being target second audio section for the audio section to be adjusted；
It, will be described if the temporal information of audio section to be adjusted is Chong Die with the temporal information of first audio section of target
The not corresponding audio section of lap in audio section to be adjusted, is determined as the target second audio section.
In one embodiment, the processor 1001 is being executed according to first text information of target identification and institute
When stating the corresponding first type of teaching information of the first audio section of target, following steps are specifically executed：
First text information of target is inputted into the first textual classification model, is obtained and first audio section of target
Matched teaching means information；
First text information of target is inputted into the second textual classification model, is obtained and first audio section of target
Matched content of courses structure；
By the corresponding teaching means information of first audio section of target and the corresponding teaching of the first audio section of the target
Content structure is determined as the first type of teaching information corresponding with first audio section of target.
In one embodiment, the processor 1001 also executes following steps：
Target second audio section by audio types for single student's type or more people student's types is converted to target second
Text information, by target first text information and the target second text envelope adjacent with second text information of target
Breath is input in first textual classification model, is obtained and the matched teaching means information of the target second audio section；
By target first text information and the target second text information adjacent with second text information of target
It is input in second textual classification model, obtains and the matched content of courses structure of the target second audio section；
By the corresponding teaching means information of the target second audio section and the corresponding teaching of the target second audio section
Content structure is determined as the second type of teaching information corresponding with the target second audio section.
In one embodiment, the processor 1001 also executes following steps：
Obtain the index model to match with the classroom environment；The index model is the classroom constructed by teaching notes
In knowledge-point models；
The knowledge point text in the index model is extracted, in first text information of target, determination is known with described
Know the object time stamp of the text information of point text matches, and the position of the object time stamp in first audio-frequency information
Add Keyword Tag；Text in the Keyword Tag is identical as the knowledge point text；
When getting searching keyword, retrieved in first audio-frequency information matched with the searching keyword
Keyword Tag, as target keywords label；
In first audio-frequency information, audio-frequency information corresponding with the timestamp of the target keywords label is exported,
And determine the curriculum information of audio-frequency information corresponding with the target keywords label.
In one embodiment, the processor 1001 is being executed according to the first type of teaching information, the target
The corresponding temporal information of first audio section, the corresponding temporal information of the target second audio section, the target second audio section
Corresponding audio types and audio frequency characteristics, it is specific to execute following step when counting the key link information in the classroom environment
Suddenly：
According to the first type of teaching information, the corresponding temporal information of the first audio section of the target, teacher's row is generated
To analyze result；
According to the second type of teaching information, the corresponding temporal information of the target second audio section, the target
The corresponding audio types of two audio sections and audio frequency characteristics generate students ' behavior and analyze result；
Teacher's behavioural analysis result and students ' behavior analysis result are determined as the pass in the classroom environment
Key link information.
In one embodiment, the processor 1001 is being executed according to the first type of teaching information, the target
First audio section corresponding temporal information specifically executes following steps when generating teacher's behavioural analysis result：
According in the first type of teaching information teaching means information and the first audio section of the target it is corresponding
Temporal information counts the total duration with the first audio section of target of identical teaching means information, obtains each teaching means letter
Cease corresponding the first parameter of teacher's behavioural analysis；
According to the teaching hand content structure and the first audio section of target correspondence in the first type of teaching information
Temporal information, count have identical content of courses structure the first audio section of target total duration, obtain each content of courses
Corresponding the second parameter of teacher's behavioural analysis of structure；
According to the teaching means information in the first type of teaching information, counting has identical teaching means information
The quantity of the first segment of target obtains the corresponding teacher's behavioural analysis third parameter of each teaching means information；
According to the content of courses structure in the first type of teaching information, counting has identical content of courses structure
The quantity of the first segment of target obtains corresponding the 4th parameter of teacher's behavioural analysis of each content of courses structure；
According to first parameter of teacher's behavioural analysis, the second parameter of teacher's behavioural analysis, teacher's behavioural analysis third ginseng
Several, described the 4th parameter of teacher's behavioural analysis, generates teacher's behavioural analysis result.
In one embodiment, the processor 1001 is being executed according to the second type of teaching information, the target
The corresponding temporal information of second audio section, the corresponding audio types of the target second audio section and audio frequency characteristics generate student
When behavioural analysis result, following steps are specifically executed：
The quantity with the target second audio section of identical audio frequency characteristics and audio types is counted, as students ' behavior point
Analyse the first parameter；
According in the second type of teaching information teaching means information and the target second audio section it is corresponding
Temporal information counts the total duration with the target second audio section of identical teaching means information, obtains each teaching means letter
It ceases corresponding students ' behavior and analyzes the second parameter；
According in the second type of teaching information content of courses structure and the target second audio section it is corresponding
Temporal information counts the total duration with the target second audio section of identical content of courses structure, obtains each content of courses knot
The corresponding students ' behavior of structure analyzes third parameter；
According to the teaching means information in the second type of teaching information, counting has identical teaching means information
The quantity of the second segment of target obtains the corresponding students ' behavior of each teaching means information and analyzes the 4th parameter；
According to the content of courses structure in the second type of teaching information, counting has identical content of courses structure
The quantity of the second segment of target obtains the corresponding students ' behavior of each content of courses structure and analyzes the 5th parameter；
The first parameter is analyzed according to the students ' behavior, the students ' behavior analyzes the second parameter, the students ' behavior point
It analyses third parameter, the 4th parameter of students ' behavior analysis, the students ' behavior and analyzes the 5th parameter, generate student's row
To analyze result.
The first audio-frequency information and second collected by obtaining the different radio reception device in two positions in classroom environment
Audio-frequency information；The audio section for belonging to teacher's type is searched in the first audio-frequency information as the first audio section of target, in the second sound
The nonoverlapping audio section of temporal information with the first audio section of target is searched in frequency information as target second audio section；By mesh
It marks the first audio section and is converted to the first text information of target, according to the identification of the first text information of target and the first audio section of target
Corresponding first type of teaching information；According to the first type of teaching information, the temporal information of the first audio section of target, target
The audio types and audio frequency characteristics of two audio sections count the key link information in classroom environment.Due to audio-frequency information
During reason is to obtain classroom key link information, without manually participating in automatically searching target sound frequency range and know
The type of teaching information of other target sound frequency range, then the key link information in programming count classroom environment, so can to avoid because
Manual analysis classroom key link information and bring tedious steps, to improve the efficiency of analysis Classroom instruction quality.Meanwhile it adopting
With the alternate analysis between the first audio-frequency information and the second audio-frequency information, can rationally, accurately analyze Classroom instruction quality, and build
Knowledge point in vertical teaching notes can quickly position the Keywords matching with input to the index relative between the first audio-frequency information
Audio-frequency information, improve classroom environment in audio recall precision.
In addition, it need to be noted that be：The embodiment of the invention also provides a kind of computer storage medium, and it is described
Computer program performed by the apparatus for processing multimedia data 1 being mentioned above is stored in computer storage medium, and described
Computer program includes program instruction, when the processor executes described program instruction, is able to carry out Fig. 1 a to Fig. 5 above
To the description of the multimedia data processing method in corresponding embodiment, therefore, will no longer repeat here.In addition, right
It is described using the beneficial effect of same procedure, is also no longer repeated.It is real for computer storage medium according to the present invention
Undisclosed technical detail in example is applied, the description of embodiment of the present invention method is please referred to.
The term used in the embodiment of the present application is only and to be not intended to limit merely for for the purpose of describing particular embodiments
The application processed.The embodiment of the present application and the "an" of singular used in the attached claims, " described " and
"the" is also intended to including most forms, unless the context clearly indicates other meaning.It is also understood that used herein
Term "and/or" refers to and includes that one or more associated any or all of project listed may combine.
Through the above description of the embodiments, it is apparent to those skilled in the art that, for description
It is convenienct and succinct, only the example of the division of the above functional modules, in practical application, it can according to need and incite somebody to action
Above-mentioned function distribution is completed by different functional modules, i.e., the internal structure of device is divided into different functional modules, with complete
At all or part of function described above.The device of foregoing description and the specific work process of unit can refer to aforementioned
Corresponding process in embodiment of the method, details are not described herein.
It, can also be in addition, each functional unit in each embodiment of the application can integrate in one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, the technical solution essence of the application
On all or part of the part that contributes to existing technology or the technical solution can be with the shape of software product in other words
Formula embodies, which is stored in a storage medium, including some instructions are used so that a calculating
Machine equipment (can be personal computer, server or the network equipment etc.) or processor (processor) execute the application
The all or part of the steps of each embodiment the method.And storage medium above-mentioned includes：USB flash disk, read-only is deposited mobile hard disk
Reservoir (Read Only Memory；Hereinafter referred to as：ROM), random access memory (Random Access Memory；Below
Referred to as：RAM), the various media that can store program code such as magnetic or disk.
The above, the only specific embodiment of the application, but the protection scope of the application is not limited thereto, and is appointed
What those familiar with the art within the technical scope of the present application, can easily think of the change or the replacement, answer
Cover within the scope of protection of this application.Therefore, the protection scope of the application should be with the scope of protection of the claims
It is quasi-.
Claims (20)
1. a kind of multimedia data processing method, which is characterized in that including：
Obtain the first audio-frequency information and the second audio-frequency information belonged in classroom environment；First audio-frequency information and described second
Audio-frequency information is that the radio reception device for being located at different location by two respectively collects；
The audio section for belonging to teacher's type is searched in first audio-frequency information, as the first audio section of target, and described
Target second audio section is searched in second audio-frequency information；The target second audio section and first audio section of target are right respectively
The temporal information answered does not overlap；
First audio section of target is converted into the first text information of target, according to first text information of target identification with
The corresponding first type of teaching information of first audio section of target；
According to the first type of teaching information, the corresponding temporal information of the first audio section of the target, the second sound of the target
The corresponding temporal information of frequency range, the corresponding audio types of the target second audio section and audio frequency characteristics, count the classroom ring
Key link information in border；The key link information includes teacher's behavioural analysis result and students ' behavior analysis result.
2. the method according to claim 1, wherein described search in first audio-frequency information belongs to teacher
The audio section of type, as the first audio section of target, including：
According to audio frequency characteristics by the first audio-frequency information cutting be multiple the first audio sections of unit；Each the first audio section of unit
It for the audio section comprising continuous sound or is mute audio section；
Each first audio section of unit is input in audio classification model, respectively identification and each first sound of unit
The audio types of frequency band match；
It is the first audio section of unit of teacher's type by the audio types in first audio-frequency information, is determined as
First audio section of target.
3. according to the method described in claim 2, it is characterized in that, described search target second in second audio-frequency information
Audio section, including：
According to audio frequency characteristics by the second audio-frequency information cutting be multiple the second audio sections of unit；Each the second audio section of unit is packet
Audio section containing continuous sound is mute audio section；
Each second audio section of unit is input in the audio classification model, respectively identification and each unit the
The matched audio types of two audio sections；
It is not the second audio section of unit of teacher's type by the audio types in second audio-frequency information, as to be adjusted
Audio section；
If the temporal information of audio section to be adjusted and the temporal information of first audio section of target be not be overlapped, it is determined that be described
If audio section to be adjusted is target second audio section；
If the temporal information of audio section to be adjusted is Chong Die with the temporal information of first audio section of target, will be described to be adjusted
The not corresponding audio section of lap in audio section, is determined as the target second audio section.
4. according to the method described in claim 3, it is characterized in that, described according to first text information of target identification and institute
The corresponding first type of teaching information of the first audio section of target is stated, including：
First text information of target is inputted into the first textual classification model, is obtained matched with first audio section of target
Teaching means information；
First text information of target is inputted into the second textual classification model, is obtained matched with first audio section of target
Content of courses structure；
By the corresponding teaching means information of first audio section of target and the corresponding content of courses of the first audio section of the target
Structure determination is the first type of teaching information corresponding with first audio section of target.
5. according to the method described in claim 4, it is characterized in that, further including：
Target second audio section by audio types for single student's type or more people student's types is converted to the second text of target
Information inputs target first text information adjacent with second text information of target and the second text information of the target
To in first textual classification model, obtain and the matched teaching means information of the target second audio section；
Target first text information adjacent with second text information of target and the second text information of the target are inputted
To in second textual classification model, obtain and the matched content of courses structure of the target second audio section；
By the corresponding teaching means information of the target second audio section and the corresponding content of courses of the target second audio section
Structure determination is the second type of teaching information corresponding with the target second audio section.
6. the method according to claim 1, wherein further including：
Obtain the index model to match with the classroom environment；The index model is knowing in the classroom constructed by teaching notes
Know point model；
The knowledge point text in the index model is extracted, in first text information of target, the determining and knowledge point
The object time of the text information of text matches stabs, and the position of the object time stamp in first audio-frequency information is added and closed
Key word label；Text in the Keyword Tag is identical as the knowledge point text；
When getting searching keyword, retrieval and the matched keyword of the searching keyword in first audio-frequency information
Label, as target keywords label；
In first audio-frequency information, audio-frequency information corresponding with the timestamp of the target keywords label is exported, and really
The curriculum information of fixed audio-frequency information corresponding with the target keywords label.
7. according to the method described in claim 5, it is characterized in that, described according to the first type of teaching information, the mesh
Mark the corresponding temporal information of the first audio section, the corresponding temporal information of the target second audio section, the target second audio
The corresponding audio types of section and audio frequency characteristics, count the key link information in the classroom environment, including：
According to the first type of teaching information, the corresponding temporal information of the first audio section of the target, teacher's behavior point is generated
Analyse result；
According to the second type of teaching information, the corresponding temporal information of the target second audio section, the second sound of the target
The corresponding audio types of frequency range and audio frequency characteristics generate students ' behavior and analyze result；
Teacher's behavioural analysis result and students ' behavior analysis result are determined as the crucial ring in the classroom environment
Save information.
8. the method according to the description of claim 7 is characterized in that described according to the first type of teaching information, the mesh
The corresponding temporal information of the first audio section is marked, generates teacher's behavioural analysis as a result, including：
According to the teaching means information and the corresponding time letter of the first audio section of the target in the first type of teaching information
Breath counts the total duration with the first audio section of target of identical teaching means information, obtains each teaching means information difference
Corresponding the first parameter of teacher's behavioural analysis；
According in the first type of teaching information teaching hand content structure and the first audio section of the target corresponding time
Information counts the total duration with the first audio section of target of identical content of courses structure, obtains each content of courses structure point
Not corresponding the second parameter of teacher's behavioural analysis；
According to the teaching means information in the first type of teaching information, the target the with identical teaching means information is counted
The quantity of one segment obtains the corresponding teacher's behavioural analysis third parameter of each teaching means information；
According to the content of courses structure in the first type of teaching information, the target the with identical content of courses structure is counted
The quantity of one segment obtains corresponding the 4th parameter of teacher's behavioural analysis of each content of courses structure；
According to first parameter of teacher's behavioural analysis, the second parameter of teacher's behavioural analysis, teacher's behavioural analysis third parameter, institute
The 4th parameter of teacher's behavioural analysis is stated, teacher's behavioural analysis result is generated.
9. the method according to the description of claim 7 is characterized in that described according to the second type of teaching information, the mesh
The corresponding temporal information of the second audio section, the corresponding audio types of the target second audio section and audio frequency characteristics are marked, generates and learns
Raw behavioural analysis is as a result, include：
The quantity with the target second audio section of identical audio frequency characteristics and audio types is counted, as students ' behavior analysis first
Parameter；
According to the teaching means information and the corresponding time letter of the target second audio section in the second type of teaching information
Breath counts the total duration with the target second audio section of identical teaching means information, obtains each teaching means information difference
Corresponding students ' behavior analyzes the second parameter；
According to the content of courses structure and the corresponding time letter of the target second audio section in the second type of teaching information
Breath counts the total duration with the target second audio section of identical content of courses structure, obtains each content of courses structure difference
Corresponding students ' behavior analyzes third parameter；
According to the teaching means information in the second type of teaching information, the target the with identical teaching means information is counted
The quantity of two segments obtains the corresponding students ' behavior of each teaching means information and analyzes the 4th parameter；
According to the content of courses structure in the second type of teaching information, the target the with identical content of courses structure is counted
The quantity of two segments obtains the corresponding students ' behavior of each content of courses structure and analyzes the 5th parameter；
The first parameter is analyzed according to the students ' behavior, the students ' behavior analyzes the second parameter, students ' behavior analysis the
Three parameters, the students ' behavior analyze the 4th parameter, the students ' behavior analyzes the 5th parameter, generate the students ' behavior analysis
As a result.
10. a kind of apparatus for processing multimedia data, which is characterized in that including：
First obtains module, for obtaining the first audio-frequency information and the second audio-frequency information that belong in classroom environment；Described first
Audio-frequency information and second audio-frequency information are that the radio reception device for being located at different location by two respectively collects；
First searching module belongs to the audio section of teacher's type for searching in first audio-frequency information, as target the
One audio section；
Second searching module, for searching target second audio section in second audio-frequency information；The target second audio
Section temporal information corresponding with first audio section of target does not overlap；
Conversion module, for first audio section of target to be converted to the first text information of target；
Identification module, for according to target text information identification the first teaching class corresponding with first audio section of target
Type information；
Statistical module, for according to the first type of teaching information, the corresponding temporal information of the first audio section of the target, institute
The corresponding temporal information of target second audio section, the corresponding audio types of the target second audio section and audio frequency characteristics are stated, are united
Count the key link information in the classroom environment；The key link information includes teacher's behavioural analysis result and students ' behavior
Analyze result.
11. device according to claim 10, which is characterized in that first searching module, including：
First cutting unit is used to according to audio frequency characteristics be multiple the first audio sections of unit by the first audio-frequency information cutting；
Each the first audio section of unit is the audio section comprising continuous sound or is mute audio section；
First recognition unit, for each first audio section of unit to be input in audio classification model, respectively identification with
The matched audio types of first audio section of each unit；
First determination unit, for by unit that the audio types in first audio-frequency information are teacher's type
One audio section is determined as first audio section of target.
12. device according to claim 11, which is characterized in that second searching module, including：
Second cutting unit is used to according to audio frequency characteristics be multiple the second audio sections of unit by the second audio-frequency information cutting；Each
The second audio section of unit is the audio section comprising continuous sound or is mute audio section；
Second recognition unit is known respectively for each second audio section of unit to be input in the audio classification model
Not with the matched audio types of each second audio section of unit；
Second determination unit, for not being the unit second of teacher's type by the audio types in second audio-frequency information
Audio section, as audio section to be adjusted；
Second determination unit, if being also used to the temporal information of audio section to be adjusted and the time of first audio section of target
Information is not overlapped, it is determined that if being target second audio section for the audio section to be adjusted；
Second determination unit, if being also used to the temporal information of audio section to be adjusted and the time of first audio section of target
Information overlap is determined as the target second audio then by the not corresponding audio section of lap in the audio section to be adjusted
Section.
13. device according to claim 12, which is characterized in that the identification module, including：
Acquiring unit obtains and the target for first text information of target to be inputted the first textual classification model
The matched teaching means information of one audio section；
The acquiring unit obtains and the mesh for first text information of target to be inputted the second textual classification model
Mark the matched content of courses structure of the first audio section；
Third determination unit is used for the corresponding teaching means information of the first audio section of target and the first audio of the target
The corresponding content of courses structure determination of section is the first type of teaching information corresponding with first audio section of target.
14. device according to claim 13, which is characterized in that further include：
The conversion module is also used to audio types be single student's type or the target second audio section of more people student's types
The second text information of target is converted to, by target first text information adjacent with second text information of target and the mesh
It marks the second text information to be input in first textual classification model, obtain and the matched teaching of target second audio section
Means Information；
First obtains module, is also used to target first text information adjacent with second text information of target and the mesh
It marks the second text information to be input in second textual classification model, obtain and the matched teaching of target second audio section
Content structure；
Determining module, for the corresponding teaching means information of the target second audio section and the target second audio section is right
The content of courses structure determination answered is the second type of teaching information corresponding with the target second audio section.
15. device according to claim 10, which is characterized in that further include：
Second obtains module, for obtaining the index model to match with the classroom environment；The index model is to pass through religion
Knowledge-point models in the classroom of case building；
Adding module in first text information of target, is determined for extracting the knowledge point text in the index model
It is stabbed with the object time of the text information of the knowledge point text matches, and the object time stamp in first audio-frequency information
Position add Keyword Tag；Text in the Keyword Tag is identical as the knowledge braille text；
Retrieval module, for when getting searching keyword, retrieval and the inquiry to be crucial in first audio-frequency information
The matched Keyword Tag of word, as target keywords label；
Output module, for exporting corresponding with the timestamp of the target keywords label in first audio-frequency information
Audio-frequency information, and determine the curriculum information of audio-frequency information corresponding with the target keywords label.
16. device according to claim 14, which is characterized in that the statistical module, including：
First generation unit, for being believed according to the first type of teaching information, the first audio section of the target corresponding time
Breath generates teacher's behavioural analysis result；
Second generation unit, for being believed according to the second type of teaching information, the target second audio section corresponding time
Breath, the corresponding audio types of the target second audio section and audio frequency characteristics generate students ' behavior and analyze result；
4th determination unit, for teacher's behavioural analysis result and students ' behavior analysis result to be determined as the class
Key link information in hall environment.
17. device according to claim 16, which is characterized in that first generation unit, including：
First statistics subelement, for according to the teaching means information and the target the in the first type of teaching information
The corresponding temporal information of one audio section counts the total duration with the first audio section of target of identical teaching means information, obtains
Corresponding the first parameter of teacher's behavioural analysis of each teaching means information；
It is described first statistics subelement, be also used to according in the first type of teaching information teaching hand content structure and institute
State the corresponding temporal information of the first audio section of target, count first audio section of target with identical content of courses structure it is total when
It is long, obtain corresponding the second parameter of teacher's behavioural analysis of each content of courses structure；
The first statistics subelement is also used to according to the teaching means information in the first type of teaching information, statistics tool
There is the quantity of the first segment of target of identical teaching means information, obtains the corresponding teacher's behavior of each teaching means information
Analyze third parameter；
The first statistics subelement is also used to according to the content of courses structure in the first type of teaching information, statistics tool
There is the quantity of the first segment of target of identical content of courses structure, obtains the corresponding teacher's behavior of each content of courses structure
Analyze the 4th parameter；
First generates subelement, for according to first parameter of teacher's behavioural analysis, the second parameter of teacher's behavioural analysis, teacher
Behavioural analysis third parameter, the 4th parameter of teacher's behavioural analysis, generate teacher's behavioural analysis result.
18. device according to claim 16, which is characterized in that second generation unit, including：
Second statistics subelement, for counting the quantity of the target second audio section with identical audio frequency characteristics and audio types,
The first parameter is analyzed as students ' behavior；
The second statistics subelement is also used to according to teaching means information in the second type of teaching information and described
The corresponding temporal information of target second audio section, count have identical teaching means information target second audio section it is total when
It is long, it obtains the corresponding students ' behavior of each teaching means information and analyzes the second parameter；
The second statistics subelement is also used to according to content of courses structure in the second type of teaching information and described
The corresponding temporal information of target second audio section, count have identical content of courses structure target second audio section it is total when
It is long, obtain the corresponding students ' behavior analysis third parameter of each content of courses structure；
The second statistics subelement is also used to according to the teaching means information in the second type of teaching information, statistics tool
There is the quantity of the second segment of target of identical teaching means information, obtains the corresponding students ' behavior of each teaching means information
Analyze the 4th parameter；
The second statistics subelement is also used to according to the content of courses structure in the second type of teaching information, statistics tool
There is the quantity of the second segment of target of identical content of courses structure, obtains the corresponding students ' behavior of each content of courses structure
Analyze the 5th parameter；
Second generate subelement, for according to the students ' behavior analyze the first parameter, the students ' behavior analyze the second parameter,
The students ' behavior analysis third parameter, the students ' behavior analyzes the 4th parameter, the students ' behavior analyzes the 5th parameter, raw
Result is analyzed at the students ' behavior.
19. a kind of electronic equipment, which is characterized in that including：Processor and memory, the processor are connected with memory,
In, the memory is configured for calling said program code for storing program code, the processor, executes such as right
It is required that the described in any item methods of 1-9.
20. a kind of computer storage medium, which is characterized in that the computer storage medium is stored with computer program, described
Computer program includes program instruction, is executed when the processor executes described program instruction such as any one of claim 1-9
The method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810549255.1A CN108920513B (en) | 2018-05-31 | 2018-05-31 | Multimedia data processing method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810549255.1A CN108920513B (en) | 2018-05-31 | 2018-05-31 | Multimedia data processing method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108920513A true CN108920513A (en) | 2018-11-30 |
| CN108920513B CN108920513B (en) | 2022-03-15 |
Family
ID=64410012
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810549255.1A Active CN108920513B (en) | 2018-05-31 | 2018-05-31 | Multimedia data processing method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108920513B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109816423A (en) * | 2018-12-14 | 2019-05-28 | 深圳壹账通智能科技有限公司 | Product programming method and server based on speech recognition |
| CN111107442A (en) * | 2019-11-25 | 2020-05-05 | 北京大米科技有限公司 | Method and device for acquiring audio and video files, server and storage medium |
| CN111210127A (en) * | 2019-12-27 | 2020-05-29 | 广东德诚科教有限公司 | Teaching data analysis method and device, computer equipment and storage medium |
| CN111681143A (en) * | 2020-04-27 | 2020-09-18 | 平安国际智慧城市科技股份有限公司 | Multi-dimensional analysis method, device, equipment and storage medium based on classroom voice |
| CN113038259A (en) * | 2021-03-05 | 2021-06-25 | 深圳市广程杰瑞科技有限公司 | Lesson quality feedback method and system for internet education |
| CN113284378A (en) * | 2021-04-12 | 2021-08-20 | 东营职业学院 | Intelligent mathematical teaching auxiliary method and device and intelligent teaching system |
| CN113360657A (en) * | 2021-06-30 | 2021-09-07 | 安徽商信政通信息技术股份有限公司 | Intelligent document distribution and handling method and device and computer equipment |
| CN113408957A (en) * | 2021-07-20 | 2021-09-17 | 北京师范大学 | Classroom teaching evaluation method based on combined empowerment method |
| CN113487213A (en) * | 2021-07-20 | 2021-10-08 | 贵州大学 | Vocational education teaching evaluation method based on big data |
| CN114005079A (en) * | 2021-12-31 | 2022-02-01 | 北京金茂教育科技有限公司 | Multimedia stream processing method and device |
| WO2024077511A1 (en) * | 2022-10-12 | 2024-04-18 | 广州视源电子科技股份有限公司 | Interaction counting method, apparatus, device, and system, and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101833584A (en) * | 2010-05-20 | 2010-09-15 | 无敌科技(西安)有限公司 | System and method for searching teaching video contents in embedded equipment |
| CN103956166A (en) * | 2014-05-27 | 2014-07-30 | 华东理工大学 | Multimedia courseware retrieval system based on voice keyword recognition |
| JP2016177045A (en) * | 2015-03-19 | 2016-10-06 | 株式会社レイトロン | Voice recognition device and voice recognition program |
| CN107492375A (en) * | 2016-06-13 | 2017-12-19 | 深圳市巨龙科教高技术股份有限公司 | A kind of classroom behavior analysis method and system |
-
2018
- 2018-05-31 CN CN201810549255.1A patent/CN108920513B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101833584A (en) * | 2010-05-20 | 2010-09-15 | 无敌科技(西安)有限公司 | System and method for searching teaching video contents in embedded equipment |
| CN103956166A (en) * | 2014-05-27 | 2014-07-30 | 华东理工大学 | Multimedia courseware retrieval system based on voice keyword recognition |
| JP2016177045A (en) * | 2015-03-19 | 2016-10-06 | 株式会社レイトロン | Voice recognition device and voice recognition program |
| CN107492375A (en) * | 2016-06-13 | 2017-12-19 | 深圳市巨龙科教高技术股份有限公司 | A kind of classroom behavior analysis method and system |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109816423A (en) * | 2018-12-14 | 2019-05-28 | 深圳壹账通智能科技有限公司 | Product programming method and server based on speech recognition |
| CN111107442B (en) * | 2019-11-25 | 2022-07-12 | 北京大米科技有限公司 | Method and device for acquiring audio and video files, server and storage medium |
| CN111107442A (en) * | 2019-11-25 | 2020-05-05 | 北京大米科技有限公司 | Method and device for acquiring audio and video files, server and storage medium |
| CN111210127A (en) * | 2019-12-27 | 2020-05-29 | 广东德诚科教有限公司 | Teaching data analysis method and device, computer equipment and storage medium |
| CN111681143A (en) * | 2020-04-27 | 2020-09-18 | 平安国际智慧城市科技股份有限公司 | Multi-dimensional analysis method, device, equipment and storage medium based on classroom voice |
| CN113038259A (en) * | 2021-03-05 | 2021-06-25 | 深圳市广程杰瑞科技有限公司 | Lesson quality feedback method and system for internet education |
| CN113038259B (en) * | 2021-03-05 | 2023-09-08 | 河南校信通教育科技有限公司 | Method and system for feeding back class quality of Internet education |
| CN113284378B (en) * | 2021-04-12 | 2022-06-21 | 东营职业学院 | Intelligent mathematical teaching auxiliary method and device and intelligent teaching system |
| CN113284378A (en) * | 2021-04-12 | 2021-08-20 | 东营职业学院 | Intelligent mathematical teaching auxiliary method and device and intelligent teaching system |
| CN113360657A (en) * | 2021-06-30 | 2021-09-07 | 安徽商信政通信息技术股份有限公司 | Intelligent document distribution and handling method and device and computer equipment |
| CN113360657B (en) * | 2021-06-30 | 2023-10-24 | 安徽商信政通信息技术股份有限公司 | Intelligent document distribution handling method and device and computer equipment |
| CN113408957A (en) * | 2021-07-20 | 2021-09-17 | 北京师范大学 | Classroom teaching evaluation method based on combined empowerment method |
| CN113487213A (en) * | 2021-07-20 | 2021-10-08 | 贵州大学 | Vocational education teaching evaluation method based on big data |
| CN113487213B (en) * | 2021-07-20 | 2022-02-01 | 贵州大学 | Vocational education teaching evaluation method based on big data |
| CN114005079A (en) * | 2021-12-31 | 2022-02-01 | 北京金茂教育科技有限公司 | Multimedia stream processing method and device |
| WO2024077511A1 (en) * | 2022-10-12 | 2024-04-18 | 广州视源电子科技股份有限公司 | Interaction counting method, apparatus, device, and system, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108920513B (en) | 2022-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108920513A (en) | A kind of multimedia data processing method, device and electronic equipment | |
| CN107222865B (en) | Communication swindle real-time detection method and system based on suspicious actions identification | |
| CN109360550B (en) | Testing method, device, equipment and storage medium of voice interaction system | |
| CN106372113B (en) | The method for pushing and system of news content | |
| CN107818164A (en) | A kind of intelligent answer method and its system | |
| CN108711420A (en) | Multilingual hybrid model foundation, data capture method and device, electronic equipment | |
| CN109815491B (en) | Answer scoring method, device, computer equipment and storage medium | |
| CN109885810A (en) | Nan-machine interrogation's method, apparatus, equipment and storage medium based on semanteme parsing | |
| CN110032630A (en) | Talk about art recommendation apparatus, method and model training equipment | |
| CN106991161A (en) | A kind of method for automatically generating open-ended question answer | |
| CN107240047A (en) | The credit appraisal procedure and device of a kind of instructional video | |
| CN108549658A (en) | A kind of deep learning video answering method and system based on the upper attention mechanism of syntactic analysis tree | |
| CN111182162A (en) | Telephone quality inspection method, device, equipment and storage medium based on artificial intelligence | |
| CN110069776B (en) | Customer satisfaction evaluation method and device and computer readable storage medium | |
| CN111681143A (en) | Multi-dimensional analysis method, device, equipment and storage medium based on classroom voice | |
| CN110164217A (en) | It a kind of online question and answer and reviews from surveying tutoring system | |
| CN110992988B (en) | Speech emotion recognition method and device based on domain confrontation | |
| CN114913729A (en) | Question selection method and device, computer equipment and storage medium | |
| CN112925888A (en) | Method and device for training question-answer response and small sample text matching model | |
| CN114186983B (en) | Video interview multidimensional scoring method, system, computer equipment and storage medium | |
| CN106708827A (en) | Quality testing method and apparatus | |
| CN117520522B (en) | Intelligent dialogue method and device based on combination of RPA and AI and electronic equipment | |
| CN112052686B (en) | Voice learning resource pushing method for user interactive education | |
| CN112767940B (en) | Voice training recognition method, system, equipment and storage medium | |
| CN113205717A (en) | Deep learning-based oral English training method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |