CN115052173A - Content analysis method and system for network short video - Google Patents
Content analysis method and system for network short video Download PDFInfo
- Publication number
- CN115052173A CN115052173A CN202210640880.3A CN202210640880A CN115052173A CN 115052173 A CN115052173 A CN 115052173A CN 202210640880 A CN202210640880 A CN 202210640880A CN 115052173 A CN115052173 A CN 115052173A
- Authority
- CN
- China
- Prior art keywords
- target
- short video
- video
- audio
- analyzing
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 30
- 238000012550 audit Methods 0.000 claims abstract description 57
- 238000013519 translation Methods 0.000 claims description 67
- 238000012937 correction Methods 0.000 claims description 48
- 238000000034 method Methods 0.000 claims description 45
- 238000012549 training Methods 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 17
- 125000004122 cyclic group Chemical group 0.000 claims description 16
- 238000000605 extraction Methods 0.000 claims description 10
- 238000012795 verification Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 description 13
- 230000006870 function Effects 0.000 description 12
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000033764 rhythmic process Effects 0.000 description 2
- 241000251468 Actinopterygii Species 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000012634 fragment Substances 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000001755 vocal effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/24—Monitoring of processes or resources, e.g. monitoring of server load, available bandwidth, upstream requests
- H04N21/2407—Monitoring of transmitted content, e.g. distribution time, number of downloads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- 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
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project 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
- G06Q50/01—Social networking
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/439—Processing of audio elementary streams
- H04N21/4394—Processing of audio elementary streams involving operations for analysing the audio stream, e.g. detecting features or characteristics in audio streams
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Signal Processing (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Computing Systems (AREA)
- Operations Research (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Data Mining & Analysis (AREA)
- Computer Networks & Wireless Communication (AREA)
- Databases & Information Systems (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a content analysis method and a system of a network short video, comprising the following steps: s1: performing machine audit on the short video in the short video platform, and acquiring a target short video qualified by the machine audit, S2: analyzing the knowledge field related to the target short video, and matching a target auditing terminal in the knowledge field, S3: generating an audit report based on the target audit terminal, S4: and correcting the target short video according to the audit report, putting the target short video into a short video platform again for playing, auditing the short video through a machine and a target audit terminal, generating an audit report after auditing, carrying out orientation adjustment on the short video according to the report, and putting the short video into the short video platform for playing, so that unnecessary loss caused by misleading audiences by non-professional knowledge is avoided.
Description
Technical Field
The invention relates to the technical field of network short videos, in particular to a content analysis method and system of a network short video.
Background
Short video is video content which is played on various new media platforms, is suitable for being watched in a mobile state and a short-time leisure state and is pushed at high frequency, the content is different from a few seconds to a few minutes, the content integrates topics such as skill sharing, humor funny, fashion trend, social hotspots, street interview, public education, advertising creativity, commercial customization and the like, and due to the fact that the content is short, the short video can be played independently and can also be a series of columns, and the short video is an indispensable leisure item in the life of people at present.
When the short video is published on a platform, the short video needs to be audited, so that misleading caused by propagation of bad information to teenagers is avoided, the current auditing of the short video adopts a mode of combining machine auditing and manual auditing, illegal videos are greatly intercepted by using the mode, but a knowledge auditing system in some professional fields is not perfect, so that some short video fish eyes which do not conform to professional knowledge are mixed, wrong guidance is provided for everybody, and serious consequences are easily caused by random simulation or learning.
In view of this, the present invention provides a method and a system for analyzing content of a network short video.
Disclosure of Invention
The content analysis method and the system of the network short video are used for auditing the short video through a machine and an authority, generating an audit report after the audit, carrying out orientation adjustment on the short video according to the report, and then putting the short video into a short video platform for playing, so that unnecessary loss caused by misleading audiences by non-professional knowledge is avoided.
The invention provides a content analysis method of a network short video, which comprises the following steps:
s1: performing machine audit on the short video in the short video platform, and acquiring a target short video qualified by the machine audit;
s2: analyzing a knowledge field related to the target short video, and matching a target auditing terminal in the knowledge field;
s3: generating an audit report based on the target audit terminal;
s4: and correcting the target short video according to the audit report, and putting the target short video into a short video platform again for playing.
In one manner that may be implemented,
s1 includes:
s11: acquiring a short video from the short video platform;
s12: acquiring flow information of the short video;
s13: obtaining the public influence degree of the short video according to the size of the flow information, matching corresponding verification precision according to the public influence degree, and performing machine verification;
s14: and acquiring the target short video qualified by machine examination.
In one manner that may be implemented,
s2 includes:
s21: extracting audio information of the target short video;
s22: extracting key information contained in the audio information, and acquiring a target knowledge field of the target short video according to the key information;
s23: determining a plurality of auditing terminals corresponding to the target knowledge field, respectively obtaining the length of an auditing queue corresponding to each auditing terminal, and determining the auditing terminal with the minimum length as a target auditing terminal.
In one manner that may be implemented,
s4 includes:
s41: analyzing problems existing in the target short video according to the audit report;
s42: analyzing the misorientation corresponding to each question to generate a correct orientation opposite to the misorientation;
s43: correcting errors in the target short video based on the correct orientation;
s44: and inputting the target short video subjected to error correction to the short video platform again.
In one manner that may be implemented,
s22 includes:
s221: acquiring audio information, and analyzing the sound frequency of the audio information;
selecting a target window function corresponding to the window width according to the sound frequency, and performing short-time Fourier transform on the audio information by using the target window function to generate a spectrogram;
s222: performing cyclic training on the spectrogram, and generating a plurality of audio sequences according to a training result;
carrying out differential processing on the audio sequence to obtain a time domain mask of the audio information;
extracting background audio from the audio information according to the time domain mask, and regarding the residual audio as human voice audio;
s223: performing homophonic translation on the human voice audio, skipping over sub-audio with ambiguity in the process of the homophonic translation, and generating a translation result;
obtaining a plurality of to-be-selected translation sub-results corresponding to the sub-audio;
respectively inputting each translation sub-result to be selected into the translation result, analyzing the coordination degree of each translation sub-result to be selected in the translation result, and extracting the optimal translation sub-result to be selected;
inputting the translation sub-result to be selected into the translation result to generate a voice subtitle;
analyzing the semantics of the voice subtitles and extracting semantic keywords from the voice subtitles;
s224: inquiring the name of the background audio, and analyzing the music attribute of the background audio according to the name;
s225: the music attribute and the semantic keyword are taken as key information;
acquiring a video context of the target short video according to the music attribute, and analyzing a matchable knowledge field of the target short video;
the semantic keywords are sequentially input into each matchable knowledge field for keyword matching, and the matching degree of the semantic keywords in each matchable knowledge field is analyzed;
and acquiring the target knowledge field of the target short video according to the matching result.
In one manner that may be implemented,
s42 includes:
s421: analyzing the video process of the target short video, and dividing the target short video into a plurality of sub-video segments according to video process nodes;
generating a problem set according to problems in the target short video, mapping the problem set to the target short video, acquiring a sub-video segment to be tuned corresponding to each problem, and generating a sub-video segment set to be tuned;
analyzing the corresponding error orientation of each sub-video segment to be tuned according to the problems;
s422: acquiring a target knowledge field of the target short video, extracting a plurality of correct orientations contained in the target knowledge field, and respectively generating corresponding orientation adjusting networks according to the correct orientations;
inputting the sub-video segment to be adjusted into each orientation adjusting network respectively to carry out first orientation adjustment, and obtaining an adjusting result of each orientation adjusting network;
s423: analyzing the distortion degree of the adjusted video clip in each adjustment result, and extracting a target adjustment result with the minimum distortion degree;
analyzing the deviation degree of the video orientation according to the correct target orientation corresponding to the target adjustment result, and generating a strength correction parameter;
acquiring a target orientation adjusting network corresponding to the target adjusting result;
correcting the adjusting force of the target orientation adjusting network for the first time according to the force correction parameter to obtain a first correction result;
generating storage correction parameters according to the number of the sub-video segments;
performing second correction on the storage space of the target orientation adjusting network according to the storage correction parameters to obtain a second correction result;
obtaining the corrected target orientation adjusting network according to the first correction result and the second correction result;
s424: inputting the sub-video segments into the corrected target orientation adjustment network respectively, and adjusting the error orientation corresponding to each sub-video segment to a target correct orientation;
and recording the adjustment modes of the video segments to be adjusted in the adjustment process, generating a correct orientation adjustment list, and transmitting the correct orientation adjustment list to a specified correction platform for displaying.
A content analysis system for network short videos, comprising:
the auditing and extracting module is used for performing machine auditing on the short videos in the short video platform and acquiring target short videos qualified by the machine auditing;
the analysis matching module is used for analyzing the knowledge field related to the target short video and matching a target auditing terminal in the knowledge field;
the analysis execution module is used for generating an audit report based on the target audit terminal;
and the error correction execution module is used for correcting the target short video according to the audit report and putting the target short video into a short video platform again for playing.
In one manner that may be implemented,
the audit extraction module comprises:
a first extraction unit, configured to obtain a short video in the short video platform;
the second extraction unit is used for acquiring the flow information of the short video;
the transmission unit is used for acquiring the public influence degree of the short video according to the size of the flow information, matching corresponding auditing precision according to the public influence degree and performing machine auditing;
and the screening unit is used for acquiring the short video of the machine-audited qualified target.
In one manner that may be implemented,
the analysis matching module comprises:
the acquisition unit is used for extracting the audio information of the target short video;
the analysis unit is used for extracting key information contained in the audio information and acquiring a target knowledge field of the target short video according to the key information;
and the matching unit is used for determining a plurality of auditing terminals corresponding to the target knowledge field, respectively acquiring the length of an auditing queue corresponding to each auditing terminal, and determining the auditing terminal with the minimum length as the target auditing terminal.
In one manner that may be implemented,
the matching unit includes:
the audio processing subunit is used for acquiring audio information and analyzing the sound frequency of the audio information;
selecting a target window function corresponding to the window width according to the sound frequency, and performing short-time Fourier transform on the audio information by using the target window function to generate a spectrogram;
the audio training subunit is used for performing cyclic training on the spectrogram and generating a plurality of audio sequences according to a training result;
carrying out differential processing on the audio sequence to obtain a time domain mask of the audio information;
extracting background audio from the audio information according to the time domain mask, and regarding the residual audio as human voice audio;
the semantic analysis subunit is used for carrying out homophonic translation on the human voice audio, skipping over sub audio with ambiguity in the process of the homophonic translation and generating a translation result;
obtaining a plurality of to-be-selected translation sub-results corresponding to the sub-audio;
respectively inputting each translation sub-result to be selected into the translation result, analyzing the coordination degree of each translation sub-result to be selected in the translation result, and extracting the optimal translation sub-result to be selected;
inputting the translation sub-result to be selected into the translation result to generate a voice caption;
analyzing the semantics of the voice subtitles and extracting semantic keywords from the voice subtitles;
the query analysis subunit is used for querying the name of the background audio and analyzing the music attribute of the background audio according to the name;
a domain matching subunit, configured to regard the music attribute and the semantic keyword as key information;
acquiring a video context of the target short video according to the music attribute, and analyzing a matchable knowledge field of the target short video;
the semantic keywords are sequentially input into each matchable knowledge field for keyword matching, and the matching degree of the semantic keywords in each matchable knowledge field is analyzed;
and acquiring the target knowledge field of the target short video according to the matching result.
Compared with the prior art, the invention has the following beneficial effects:
in order to overcome the defects of traditional auditing, machine auditing is firstly carried out on short videos in a short video platform, then the target short videos qualified in the machine auditing are transmitted to an authoritative auditing end for auditing, an auditing report is generated, finally error correction is carried out on the target short videos according to the report, and then the target short videos are displayed in front of the public, and authoritative and real video guidance is provided for the public.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
fig. 1 is a schematic flow chart illustrating a method for analyzing content of a network short video according to an embodiment of the present invention;
fig. 2 is a schematic flowchart of a content analysis method S1 for online short videos according to an embodiment of the present invention;
fig. 3 is a flowchart illustrating a content analysis method S2 for network short video according to an embodiment of the present invention;
fig. 4 is a schematic diagram illustrating a system for analyzing content of a network short video according to an embodiment of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
Example 1
A method for analyzing content of a network short video, as shown in fig. 1, includes:
s1: performing machine audit on the short video in the short video platform, and acquiring a target short video qualified by the machine audit;
s2: analyzing a knowledge field related to the target short video, and matching a target auditing terminal in the knowledge field;
s3: generating an audit report based on the target audit terminal;
s4: and correcting the target short video according to the audit report, and putting the target short video into a short video platform again for playing.
In the example, the machine audit represents the process of performing non-intelligent audit on the target short video, and short videos meeting legal regulations can be screened out in the process;
in this example, the process of generating the audit report may be: professional examiners in the knowledge field intelligently examine the target short video, and the authenticity of the content related to the target short video can be analyzed in the process;
in this example, the audit report indicates a report generated after a professional reviewer employee performs an authoritative audit on the target short video.
The working principle and the beneficial effects of the technical scheme are as follows: in order to overcome the defects of traditional auditing, machine auditing is firstly carried out on short videos in a short video platform, then target short videos qualified in machine auditing are transmitted to an authoritative auditing end for auditing, an auditing report is generated, finally error correction is carried out on the target short videos according to the report, the target short videos are displayed in front of the public, authoritative and real video guidance is provided for the public, the short videos are audited through a machine and a target auditing terminal, an auditing report is generated after auditing, orientation adjustment is carried out on the short videos according to the report, the short videos are put into the short video platform for playing, and unnecessary loss caused by misleading of non-professional knowledge to audiences is avoided.
Example 2
On the basis of embodiment 1, in the method for analyzing content of a network short video, as shown in fig. 2, S1 includes:
s11: acquiring a short video from the short video platform;
s12: acquiring flow information of the short video;
s13: obtaining the public influence degree of the short video according to the size of the flow information, matching corresponding verification precision according to the public influence degree, and performing machine verification;
s14: and acquiring the target short video qualified by machine examination.
In this example, the traffic information indicates the number of plays and the number of praise of the short video;
in this example, the larger the traffic information of the short video, the higher the audit progress.
The working principle and the beneficial effects of the technical scheme are as follows: in order to improve the auditing efficiency, the flow information of the short video is acquired before auditing, the degree of influence of the video on the public is analyzed, then machine auditing is carried out by matching with different auditing precisions, only the short video with qualified auditing is reserved, the basis is taken for subsequent authoritative auditing, the authoritative auditing efficiency is improved to a certain extent, and high-efficiency auditing is realized.
Example 3
On the basis of embodiment 1, in the method for analyzing content of a network short video, as shown in fig. 3, S2 includes:
s21: extracting audio information of the target short video;
s22: extracting key information contained in the audio information, and acquiring a target knowledge field of the target short video according to the key information;
s23: determining a plurality of auditing terminals corresponding to the target knowledge field, respectively obtaining the length of an auditing queue corresponding to each auditing terminal, and determining the auditing terminal with the minimum length as a target auditing terminal.
In this example, the audio information represents audio extracted from the target short video;
in this example, the target knowledge domain represents the knowledge domain to which the target short video belongs;
in this example, the audit queue represents the amount of stacked tasks for the audit terminal.
The working principle and the beneficial effects of the technical scheme are as follows: in order to quickly position the knowledge field of the target short video, the key information of the target short video is obtained by analyzing the audio information of the target short video, the target knowledge field to which the target short video belongs is analyzed, then the length of each audit queue in the field is used as the target short video matching audit terminal, intelligent matching is achieved, and audit progress is accelerated.
Example 4
On the basis of embodiment 1, the method for analyzing content of network short video, S4, includes:
s41: analyzing problems existing in the target short video according to the audit report;
s42: analyzing the misorientation corresponding to each question to generate a correct orientation opposite to the misorientation;
s43: correcting errors in the target short video based on the correct orientation;
s44: and inputting the target short video subjected to error correction to the short video platform again.
In this example, the misorientation represents an expressed orientation of the video segment in question in the target short video;
in this example, the correct orientation represents the standard orientation that the fragment should express.
The working principle and the beneficial effects of the technical scheme are as follows: in order to improve the quality of the video in the short video platform, the target short video is processed according to the wrong orientation to obtain the correct orientation, then the target short video is corrected to realize error correction, and then the target short video is put on the short video platform to be played.
Example 5
On the basis of embodiment 3, the method for analyzing content of network short video, S22, includes:
s221: acquiring audio information, and analyzing the sound frequency of the audio information;
selecting a target window function corresponding to the window width according to the sound frequency, and performing short-time Fourier transform on the audio information by using the target window function to generate a spectrogram;
s222: performing cyclic training on the spectrogram, and generating a plurality of audio sequences according to a training result;
carrying out differential processing on the audio sequence to obtain a time domain mask of the audio information;
extracting background audio from the audio information according to the time domain mask, and regarding the residual audio as human voice audio;
s223: carrying out simultaneous interpretation on the human voice audio, skipping over ambiguous sub-audio in the simultaneous interpretation process, and generating an interpretation result;
obtaining a plurality of to-be-selected translation sub-results corresponding to the sub-audio;
respectively inputting each translation sub-result to be selected into the translation result, analyzing the coordination degree of each translation sub-result to be selected in the translation result, and extracting the optimal translation sub-result to be selected;
inputting the translation sub-result to be selected into the translation result to generate a voice subtitle;
analyzing the semantics of the voice subtitles and extracting semantic keywords from the voice subtitles;
s224: inquiring the name of the background audio, and analyzing the music attribute of the background audio according to the name;
s225: regarding the music attribute and the semantic keyword as key information;
acquiring a video context of the target short video according to the music attribute, and analyzing a matchable knowledge field of the target short video;
the semantic keywords are sequentially input into each matchable knowledge field for keyword matching, and the matching degree of the semantic keywords in each matchable knowledge field is analyzed;
and acquiring the target knowledge field of the target short video according to the matching result.
In this example, the sound frequency represents the number of times the sound in the audio information periodically vibrates per second;
in this example, the spectrogram represents a sound spectrogram obtained by time-domain processing audio information;
in this example, the process of performing cyclic training on the spectrogram is as follows: establishing a cyclic neural network framework, and inputting the spectrogram into the cyclic neural network framework for cyclic training;
in this example, the criteria for selecting the window function are: the window width of the window function may cover the peak of the sound frequency;
in this example, the difference processing represents a process of performing the equalization processing on the sounds in two adjacent audio sequences by using the equalization method, and analyzing the ordered rhythms in the audio sequences;
in this example, the time domain mask represents the intensity of an ordered tempo in the audio sequence that is masked by human voice or other unordered sound;
in this example, the background audio represents background music of the target short video;
in this example, the optimal translation result represents a translation result that is consistent with the core meaning of the target short video and is most easily understood;
in this example, the semantic keyword represents a word or word that may summarize a human voice caption.
The working principle and the beneficial effects of the technical scheme are as follows: in order to quickly classify target short videos, firstly, Fourier transform is carried out on audio information of the target short videos to obtain a spectrogram, cyclic training is carried out on the spectrogram to generate a plurality of audio sequences, then background music and human voice in the target short videos are separated, different processing is carried out respectively, finally, key information is generated by obtaining music attributes and semantic keywords of the target short videos, domain matching is carried out on the target short videos by utilizing the key information to obtain a target knowledge domain, and the target knowledge domain is used as a basis for subsequent authoritative audit.
Example 6
On the basis of embodiment 4, the method for analyzing content of network short video, S42, includes:
s421: analyzing the video process of the target short video, and dividing the target short video into a plurality of sub-video segments according to video process nodes;
generating a problem set according to problems in the target short video, mapping the problem set to the target short video, acquiring a sub-video segment to be tuned corresponding to each problem, and generating a sub-video segment set to be tuned;
analyzing the corresponding error orientation of each sub-video segment to be tuned according to the problem;
s422: acquiring a target knowledge field of the target short video, extracting a plurality of correct orientations contained in the target knowledge field, and respectively generating corresponding orientation adjusting networks according to the correct orientations;
inputting the sub-video segment to be adjusted into each orientation adjusting network respectively to carry out first orientation adjustment, and obtaining an adjusting result of each orientation adjusting network;
s423: analyzing the distortion degree of the adjusted video clip in each adjustment result, and extracting a target adjustment result with the minimum distortion degree;
analyzing the deviation degree of the video orientation according to the target correct orientation corresponding to the target adjusting result to generate a force correction parameter;
acquiring a target orientation adjusting network corresponding to the target adjusting result;
correcting the adjusting force of the target orientation adjusting network for the first time according to the force correction parameter to obtain a first correction result;
generating storage correction parameters according to the number of the sub-video segments;
performing second correction on the storage space of the target orientation adjusting network according to the storage correction parameters to obtain a second correction result;
obtaining the corrected target orientation adjusting network according to the first correction result and the second correction result;
s424: inputting the sub-video segments into the corrected target orientation adjusting network respectively, and adjusting the corresponding wrong orientation of each sub-video segment to a target correct orientation;
and recording the adjustment modes of the video segments to be adjusted in the adjustment process, generating a correct orientation adjustment list, and transmitting the correct orientation adjustment list to a specified correction platform for displaying.
In this example, the video process represents a playing process of the target short video;
in this example, the video progress node represents a pause point and a turning point in the playing process of the target short video;
in this example, the problem set represents the set of all problems, in this embodiment the mapped pre-image set;
in this example, the sub-video segment set to be tuned represents a collection of all sub-video segments to be tuned that are matched in the target short video according to the question set, and in this embodiment, represents a mapped image set;
in this example, the misorientation represents a guide to content presented by the video segment to be tuned;
in this example, the video orientation represents a guide to all content in the destination short video;
in this example, the correct orientation represents a guide contained in the target knowledge domain;
in this example, the orientation adjustment network represents an adjustment network generated with each correct orientation as an output result, and one correct orientation corresponds to one orientation adjustment network;
in this example, the storage space represents a spatial specification of the target short video after the target orientation adjustment network stores the adjustment;
in this example, the adjustment strength represents the strength of adjusting the target short video;
in this example, the degree of deviation represents the difference between the video orientation and the target correct orientation;
in this example, the first correction represents a correction of the adjustment strength of the target orientation adjustment network;
in this example, the first correction result represents the target orientation adjustment network with the corrected adjustment strength;
in this example, the second correction represents a correction of the spatial specification of the target orientation adjustment network;
in this example, the second correction result represents the target orientation adjustment network of the corrected storage space.
The working principle and the beneficial effects of the technical scheme are as follows: in order to furthest reserve the original content of a target short video in the process of correcting the error of the target short video, firstly, a problem set is generated according to problems existing in the target short video, then sub-video segments to be adjusted are matched in the target short video, a sub-video segment set to be adjusted is generated, the error orientation of each video segment to be adjusted and the video orientation of the target short video are analyzed, then the video segments to be adjusted are input into an orientation adjusting network and adjusted to the correct orientation, then, a target orientation adjusting network is selected according to the adjusted distortion degree, in order to ensure that the orientation of each sub-video segment in the target short video is correct, the target short video is input into the target orientation adjusting network to be integrally adjusted, an orientation comparison list is generated, and a basis is made for subsequent specific adjustment.
Example 7
On the basis of embodiment 5, in the method for analyzing content of network short video, S224, extracting semantic keywords from the human voice subtitles, includes:
step A: acquiring the voice captions, and dividing the voice captions into sentence breaks according to punctuations of the voice captions to generate a plurality of caption sentences;
respectively carrying out word division on each subtitle sentence to generate a corresponding word set;
and B: constructing a word library based on the word set, and performing attribute classification according to the word attribute of each word;
marking a target word in the caption sentence, and replacing the target word by using the residual word in the target word attribute corresponding to the target word;
after the replacement is completed each time, calculating the similarity between the replaced caption sentence and the original caption sentence by using a formula (I);
wherein T represents the similarity between the replaced caption sentence and the original caption sentence, and T a Represents a first degree of coherence between the a-th word and the a + 1-th word in the subtitle sentence before replacement,
representing a second consistency between the a-th word and the a + 1-th word in the replaced caption sentence, and g representing the total number of words contained in the original caption sentence and the total number of words contained in the replaced caption sentence, wherein the two are the same;
extracting target words and residual words corresponding to the similarity between the replaced subtitle sentence and the original subtitle sentence within a preset similarity range according to the calculation result of the formula (I), and establishing a replacement relation;
acquiring all the replacement relations, and establishing a corresponding list of similar words;
and C: calculating the importance of each word in the word library in the voice caption based on a formula (II) according to the corresponding list of the similar words;
wherein L is i Representing the importance of the ith word in the word library in the voice caption, f i Representing the number of times the ith word appears in the word bank, f x Representing the times of appearance of the x-th word with similar semanteme to the i-th word in the corresponding list of similar words in the word library, k representing the number of all words with similar semanteme to the i-th word in the word library, t x Representing the word meaning similarity between the xth word and the ith word, and D representing the number of words in the word library;
step D: and extracting important words with the importance within a preset importance range, and regarding the important words as semantic keywords.
The working principle and the beneficial effects of the technical scheme are as follows: in order to realize keyword extraction, a word library is established, words in caption sentences are replaced by words with the same attribute, the similarity of different words in the word library is analyzed, the importance of each word in the voice caption is calculated, whether the word belongs to the keyword or not is reflected laterally according to the similarity of the words, semantic keywords can be obtained by the method, extraction work is completed, and a foundation is laid for subsequent field matching.
Example 8
A content analysis system for network short videos, comprising:
the auditing and extracting module is used for performing machine auditing on the short videos in the short video platform and acquiring target short videos qualified by the machine auditing;
the analysis matching module is used for analyzing the knowledge field related to the target short video and matching a target auditing terminal in the knowledge field;
the analysis execution module is used for generating an audit report based on the target audit terminal;
and the error correction execution module is used for correcting the target short video according to the audit report and putting the target short video into a short video platform again for playing.
The working principle and the beneficial effects of the technical scheme are as follows: the short video and the machines are checked firstly, then authority checking is carried out, and finally the purpose of screening and error correction is achieved, and the authenticity and the safety of the video are guaranteed.
Example 9
On the basis of embodiment 8, the system for analyzing content of network short videos includes:
the first extraction unit is used for acquiring a short video from the short video platform;
the second extraction unit is used for acquiring the flow information of the short video;
the transmission unit is used for acquiring the public influence degree of the short video according to the size of the flow information, matching corresponding auditing precision according to the public influence degree and performing machine auditing;
and the screening unit is used for acquiring the short video of the machine-audited qualified target.
The working principle and the beneficial effects of the technical scheme are as follows: by performing machine audit on the short video, the spread of the junk video is greatly reduced, an audit basis is provided for authoritative audit, and the audit speed is increased.
Example 10
On the basis of embodiment 8, the content analysis system for network short videos includes:
the acquisition unit is used for extracting the audio information of the target short video;
the analysis unit is used for extracting key information contained in the audio information and acquiring a target knowledge field of the target short video according to the key information;
and the matching unit is used for determining a plurality of auditing terminals corresponding to the target knowledge field, respectively acquiring the length of an auditing queue corresponding to each auditing terminal, and determining the auditing terminal with the minimum length as a target auditing terminal.
The working principle and the beneficial effects of the technical scheme are as follows: by performing field distribution on the target short video and then submitting the target short video to the target auditing terminal, the personnel in the professional field can audit professional knowledge, the auditing quality is improved, and the gold content of the short video is ensured.
Example 11
On the basis of embodiment 8, the system for analyzing content of network short video includes:
the comparison and analysis unit is used for analyzing the problems in the target short video according to the audit report;
an orientation generating unit, configured to analyze a wrong orientation corresponding to each question, and generate a correct orientation opposite to the wrong orientation;
a correction execution unit for correcting errors of the target short video based on the correct orientation;
and the releasing unit is used for releasing the target short video after error correction to the short video platform again.
The working principle and the beneficial effects of the technical scheme are as follows: in order to ensure the quality of the video, the error correction execution unit analyzes the problems of the target short video by using the audit report, then adjusts the error orientation of the target short video, and puts the adjusted short video into the short video platform again to improve the credibility of the video.
Example 12
On the basis of embodiment 10, the content analysis system for network short videos includes:
the audio processing subunit is used for acquiring audio information and analyzing the sound frequency of the audio information;
selecting a target window function corresponding to the window width according to the sound frequency, and performing short-time Fourier transform on the audio information by using the target window function to generate a spectrogram;
the audio training subunit is used for performing cyclic training on the spectrogram and generating a plurality of audio sequences according to a training result;
carrying out differential processing on the audio sequence to obtain a time domain mask of the audio information;
extracting background audio from the audio information according to the time domain mask, and regarding the residual audio as human voice audio;
the semantic analysis subunit is used for carrying out homophonic translation on the human voice audio, skipping over sub-audio with ambiguity in the process of the homophonic translation and generating a translation result;
obtaining a plurality of to-be-selected translation sub-results corresponding to the sub-audio;
respectively inputting each translation sub-result to be selected into the translation result, analyzing the coordination degree of each translation sub-result to be selected in the translation result, and extracting the optimal translation sub-result to be selected;
inputting the translation sub-result to be selected into the translation result to generate a voice subtitle;
analyzing the semantics of the voice subtitles and extracting semantic keywords from the voice subtitles;
the query analysis subunit is used for querying the name of the background audio and analyzing the music attribute of the background audio according to the name;
a domain matching subunit, configured to regard the music attribute and the semantic keyword as key information;
acquiring the video context of the target short video according to the music attribute, and analyzing the field of matchable knowledge of the target short video;
the semantic keywords are sequentially input into each matchable knowledge field for keyword matching, and the matching degree of the semantic keywords in each matchable knowledge field is analyzed;
and acquiring the target knowledge field of the target short video according to the matching result.
In this example, the sound frequency represents the number of times the sound in the audio information periodically vibrates per second;
in this example, the spectrogram represents a sound spectrogram obtained by time-domain processing audio information;
in this example, the process of performing cyclic training on the spectrogram is as follows: establishing a cyclic neural network framework, and inputting the spectrogram into the cyclic neural network framework for cyclic training;
in this example, the criteria for selecting the window function are: the window width of the window function may cover the peak of the sound frequency;
in this example, the difference processing represents a process of performing the equalization processing on the sounds in two adjacent audio sequences by using the equalization method, and analyzing the ordered rhythms in the audio sequences;
in this example, the time domain mask represents the intensity of an ordered tempo in the audio sequence that is masked by human voice or other unordered sound;
in this example, the background audio represents background music of the target short video;
in this example, the optimal translation result represents a translation result that is consistent with the core meaning of the target short video and is most easily understood;
in this example, the semantic keyword represents a word or word that may summarize a vocal subtitle.
The working principle and the beneficial effects of the technical scheme are as follows: in order to ensure the speed and quality of auditing, it is very important to match the knowledge field for the target short video before auditing, firstly, the audio information of the target short video is subjected to Fourier transform to obtain a spectrogram, then, the spectrogram is subjected to cyclic training to generate a plurality of audio sequences, then, background music and voice in the target short video are separated, then, different processing is respectively carried out, finally, the music attribute and the semantic keyword of the target short video are obtained to generate key information, the key information is used for carrying out field matching on the target short video to obtain a target knowledge field, and thus, the method can be used as a basis for subsequent auditing.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. A content analysis method of a network short video is characterized by comprising the following steps:
s1: performing machine audit on the short video in the short video platform, and acquiring a target short video qualified by the machine audit;
s2: analyzing a knowledge field related to the target short video, and matching a target auditing terminal in the knowledge field;
s3: generating an audit report based on the target audit terminal;
s4: and correcting the target short video according to the audit report, and putting the target short video into a short video platform again for playing.
2. The method for analyzing content of network short video according to claim 1, wherein S1 comprises:
s11: acquiring a short video from the short video platform;
s12: acquiring the flow information of the short video;
s13: obtaining the public influence degree of the short video according to the size of the flow information, matching corresponding verification precision according to the public influence degree, and performing machine verification;
s14: and acquiring the target short video qualified by machine examination.
3. The method for analyzing content of network short video according to claim 1, wherein S2 comprises:
s21: extracting audio information of the target short video;
s22: extracting key information contained in the audio information, and acquiring a target knowledge field of the target short video according to the key information;
s23: determining a plurality of auditing terminals corresponding to the target knowledge field, respectively obtaining the length of an auditing queue corresponding to each auditing terminal, and determining the auditing terminal with the minimum length as a target auditing terminal.
4. The method for analyzing content of online short video as claimed in claim 1, wherein S4 comprises:
s41: analyzing problems existing in the target short video according to the audit report;
s42: analyzing the misorientation corresponding to each question to generate a correct orientation opposite to the misorientation;
s43: correcting errors in the target short video based on the correct orientation;
s44: and putting the target short video subjected to error correction into the short video platform again.
5. The method for analyzing content of network short video as claimed in claim 3, wherein the S22 comprises:
s221: acquiring audio information, and analyzing the sound frequency of the audio information;
selecting a target window function corresponding to the window width according to the sound frequency, and performing short-time Fourier transform on the audio information by using the target window function to generate a spectrogram;
s222: performing cyclic training on the spectrogram, and generating a plurality of audio sequences according to a training result;
carrying out differential processing on the audio sequence to obtain a time domain mask of the audio information;
extracting background audio from the audio information according to the time domain mask, and regarding the residual audio as human voice audio;
s223: performing homophonic translation on the human voice audio, skipping over sub-audio with ambiguity in the process of the homophonic translation, and generating a translation result;
obtaining a plurality of to-be-selected translation sub-results corresponding to the sub-audio;
respectively inputting each translation sub-result to be selected into the translation result, analyzing the coordination degree of each translation sub-result to be selected in the translation result, and extracting the optimal translation sub-result to be selected;
inputting the translation sub-result to be selected into the translation result to generate a voice subtitle;
analyzing the semantics of the voice subtitles and extracting semantic keywords from the voice subtitles;
s224: inquiring the name of the background audio, and analyzing the music attribute of the background audio according to the name;
s225: regarding the music attribute and the semantic keyword as key information;
acquiring a video context of the target short video according to the music attribute, and analyzing a matchable knowledge field of the target short video;
the semantic keywords are sequentially input into each matchable knowledge field for keyword matching, and the matching degree of the semantic keywords in each matchable knowledge field is analyzed;
and acquiring the target knowledge field of the target short video according to the matching result.
6. The method for analyzing content of network short video as claimed in claim 4, wherein the S42 comprises:
s421: analyzing the video process of the target short video, and dividing the target short video into a plurality of sub-video segments according to video process nodes;
generating a problem set according to problems in the target short video, mapping the problem set to the target short video, acquiring a sub-video segment to be tuned corresponding to each problem, and generating a sub-video segment set to be tuned;
analyzing the corresponding error orientation of each sub-video segment to be tuned according to the problems;
s422: acquiring a target knowledge field of the target short video, extracting a plurality of correct orientations contained in the target knowledge field, and respectively generating corresponding orientation adjusting networks according to the correct orientations;
inputting the sub-video segment to be adjusted into each orientation adjusting network respectively to carry out first orientation adjustment, and obtaining an adjusting result of each orientation adjusting network;
s423: analyzing the distortion degree of the adjusted video clip in each adjustment result, and extracting a target adjustment result with the minimum distortion degree;
analyzing the deviation degree of the video orientation according to the target correct orientation corresponding to the target adjusting result to generate a force correction parameter;
acquiring a target orientation adjusting network corresponding to the target adjusting result;
correcting the adjusting force of the target orientation adjusting network for the first time according to the force correction parameter to obtain a first correction result;
generating storage correction parameters according to the number of the sub-video segments;
performing second correction on the storage space of the target orientation adjusting network according to the storage correction parameters to obtain a second correction result;
obtaining the corrected target orientation adjusting network according to the first correction result and the second correction result;
s424: inputting the sub-video segments into the corrected target orientation adjusting network respectively, and adjusting the corresponding wrong orientation of each sub-video segment to a target correct orientation;
and recording the adjusting mode corresponding to each sub-video segment in the adjusting process, generating a correct orientation adjusting list, and transmitting the correct orientation adjusting list to a specified correcting platform for displaying.
7. A system for analyzing content of a network short video, comprising:
the auditing and extracting module is used for performing machine auditing on the short videos in the short video platform and acquiring target short videos qualified by the machine auditing;
the analysis matching module is used for analyzing the knowledge field related to the target short video and matching a target auditing terminal in the knowledge field;
the analysis execution module is used for generating an audit report based on the target audit terminal;
and the error correction execution module is used for correcting the target short video according to the audit report and putting the target short video into a short video platform again for playing.
8. The system for analyzing content of network short video according to claim 7, wherein the audit extraction module comprises:
the first extraction unit is used for acquiring a short video from the short video platform;
the second extraction unit is used for acquiring the flow information of the short video;
the transmission unit is used for acquiring the public influence degree of the short video according to the size of the flow information, matching corresponding auditing precision according to the public influence degree and performing machine auditing;
and the screening unit is used for acquiring the short video of the machine-audited qualified target.
9. The system for analyzing content of network short video according to claim 7, wherein the analyzing and matching module comprises:
the acquisition unit is used for extracting the audio information of the target short video;
the analysis unit is used for extracting key information contained in the audio information and acquiring a target knowledge field of the target short video according to the key information;
and the matching unit is used for determining a plurality of auditing terminals corresponding to the target knowledge field, respectively acquiring the length of an auditing queue corresponding to each auditing terminal, and determining the auditing terminal with the minimum length as the target auditing terminal.
10. The system for analyzing content of network short video according to claim 9, wherein the matching unit comprises:
the audio processing subunit is used for acquiring audio information and analyzing the sound frequency of the audio information;
selecting a target window function corresponding to the window width according to the sound frequency, and performing short-time Fourier transform on the audio information by using the target window function to generate a spectrogram;
the audio training subunit is used for performing cyclic training on the spectrogram and generating a plurality of audio sequences according to a training result;
carrying out differential processing on the audio sequence to obtain a time domain mask of the audio information;
extracting background audio from the audio information according to the time domain mask, and regarding the residual audio as human voice audio;
the semantic analysis subunit is used for carrying out homophonic translation on the human voice audio, skipping over sub audio with ambiguity in the process of the homophonic translation and generating a translation result;
obtaining a plurality of to-be-selected translation sub-results corresponding to the sub-audio;
respectively inputting each translation sub-result to be selected into the translation result, analyzing the coordination degree of each translation sub-result to be selected in the translation result, and extracting the optimal translation sub-result to be selected;
inputting the translation sub-result to be selected into the translation result to generate a voice subtitle;
analyzing the semantics of the voice subtitles and extracting semantic keywords from the voice subtitles;
the query analysis subunit is used for querying the name of the background audio and analyzing the music attribute of the background audio according to the name;
a domain matching subunit, configured to regard the music attribute and the semantic keyword as key information;
acquiring a video context of the target short video according to the music attribute, and analyzing a matchable knowledge field of the target short video;
the semantic keywords are sequentially input into each matchable knowledge field for keyword matching, and the matching degree of the semantic keywords in each matchable knowledge field is analyzed;
and acquiring the target knowledge field of the target short video according to the matching result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210640880.3A CN115052173B (en) | 2022-06-07 | 2022-06-07 | Content analysis method and system for network short video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210640880.3A CN115052173B (en) | 2022-06-07 | 2022-06-07 | Content analysis method and system for network short video |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115052173A true CN115052173A (en) | 2022-09-13 |
| CN115052173B CN115052173B (en) | 2023-08-01 |
Family
ID=83162089
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210640880.3A Active CN115052173B (en) | 2022-06-07 | 2022-06-07 | Content analysis method and system for network short video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115052173B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101035280A (en) * | 2007-04-19 | 2007-09-12 | 鲍东山 | Classified content auditing terminal system |
| US20140085480A1 (en) * | 2008-03-03 | 2014-03-27 | Videolq, Inc. | Content-aware computer networking devices with video analytics for reducing video storage and video communication bandwidth requirements of a video surveillance network camera system |
| CN105898414A (en) * | 2015-11-13 | 2016-08-24 | 乐视云计算有限公司 | Video reviewing method and system |
| CN108419091A (en) * | 2018-03-02 | 2018-08-17 | 北京未来媒体科技股份有限公司 | A kind of verifying video content method and device based on machine learning |
| CN110225373A (en) * | 2019-06-13 | 2019-09-10 | 腾讯科技（深圳）有限公司 | A kind of video reviewing method, device and electronic equipment |
| CN111753133A (en) * | 2020-06-11 | 2020-10-09 | 北京小米松果电子有限公司 | Video classification method, device and storage medium |
| CN112153397A (en) * | 2020-09-16 | 2020-12-29 | 北京达佳互联信息技术有限公司 | Video processing method, device, server and storage medium |
| CN113312504A (en) * | 2021-07-30 | 2021-08-27 | 北京远鉴信息技术有限公司 | Management method, device, equipment and medium for content audit project |
| CN113537940A (en) * | 2021-07-22 | 2021-10-22 | 北京华雨天成文化传播有限公司 | Audiovisual program content auditing and management method and system based on block chain technology |
| CN113850162A (en) * | 2021-09-10 | 2021-12-28 | 北京百度网讯科技有限公司 | Video auditing method and device and electronic equipment |
| CN113901839A (en) * | 2021-11-18 | 2022-01-07 | 平安普惠企业管理有限公司 | User video information auditing method, device, equipment and storage medium |
| CN114463674A (en) * | 2022-01-04 | 2022-05-10 | 北京快乐茄信息技术有限公司 | Video auditing method and device |
-
2022
- 2022-06-07 CN CN202210640880.3A patent/CN115052173B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101035280A (en) * | 2007-04-19 | 2007-09-12 | 鲍东山 | Classified content auditing terminal system |
| US20140085480A1 (en) * | 2008-03-03 | 2014-03-27 | Videolq, Inc. | Content-aware computer networking devices with video analytics for reducing video storage and video communication bandwidth requirements of a video surveillance network camera system |
| CN105898414A (en) * | 2015-11-13 | 2016-08-24 | 乐视云计算有限公司 | Video reviewing method and system |
| WO2017080168A1 (en) * | 2015-11-13 | 2017-05-18 | 乐视控股（北京）有限公司 | Video reviewing method and system |
| CN108419091A (en) * | 2018-03-02 | 2018-08-17 | 北京未来媒体科技股份有限公司 | A kind of verifying video content method and device based on machine learning |
| CN110225373A (en) * | 2019-06-13 | 2019-09-10 | 腾讯科技（深圳）有限公司 | A kind of video reviewing method, device and electronic equipment |
| CN111753133A (en) * | 2020-06-11 | 2020-10-09 | 北京小米松果电子有限公司 | Video classification method, device and storage medium |
| CN112153397A (en) * | 2020-09-16 | 2020-12-29 | 北京达佳互联信息技术有限公司 | Video processing method, device, server and storage medium |
| CN113537940A (en) * | 2021-07-22 | 2021-10-22 | 北京华雨天成文化传播有限公司 | Audiovisual program content auditing and management method and system based on block chain technology |
| CN113312504A (en) * | 2021-07-30 | 2021-08-27 | 北京远鉴信息技术有限公司 | Management method, device, equipment and medium for content audit project |
| CN113850162A (en) * | 2021-09-10 | 2021-12-28 | 北京百度网讯科技有限公司 | Video auditing method and device and electronic equipment |
| CN113901839A (en) * | 2021-11-18 | 2022-01-07 | 平安普惠企业管理有限公司 | User video information auditing method, device, equipment and storage medium |
| CN114463674A (en) * | 2022-01-04 | 2022-05-10 | 北京快乐茄信息技术有限公司 | Video auditing method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115052173B (en) | 2023-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110148427B (en) | Audio processing method, device, system, storage medium, terminal and server | |
| WO2021017612A1 (en) | Question and answer method and apparatus based on speech analysis, device and storage medium | |
| US8535236B2 (en) | Apparatus and method for analyzing a sound signal using a physiological ear model | |
| CN108268530B (en) | Lyric score generation method and related device | |
| US11443731B2 (en) | Systems and methods for generating synthesized speech responses to voice inputs by training a neural network model based on the voice input prosodic metrics and training voice inputs | |
| Müller et al. | Interactive fundamental frequency estimation with applications to ethnomusicological research | |
| Hou et al. | Transfer learning for improving singing-voice detection in polyphonic instrumental music | |
| Caro Repetto et al. | Comparision of the singing style of two Jingju schools | |
| CN114125506B (en) | Voice auditing method and device | |
| CN111402919B (en) | Method for identifying style of playing cavity based on multi-scale and multi-view | |
| CN115052173A (en) | Content analysis method and system for network short video | |
| Kitahara et al. | Musical instrument recognizer" instrogram" and its application to music retrieval based on instrumentation similarity | |
| CN110853457A (en) | Interactive music teaching guidance method | |
| Lukasik | Long term cepstral coefficients for violin identification | |
| Bhatia et al. | Analysis of audio features for music representation | |
| Zhang | Violin teaching improvement strategy in the context of intelligent internet of things | |
| CN114241835A (en) | Student spoken language quality evaluation method and device | |
| Gärtner | Unsupervised learning of the downbeat in drum patterns | |
| CN113038259A (en) | Lesson quality feedback method and system for internet education | |
| CN112201100A (en) | Music singing scoring system and method for evaluating artistic quality of primary and secondary schools | |
| Desblancs | Self-supervised beat tracking in musical signals with polyphonic contrastive learning | |
| Faisal et al. | ESLCE: A Dataset of Emotional Sounds from Large Crowd Events | |
| Heinz et al. | Using a physiological ear model for automatic melody transcription and sound source recognition | |
| Pham | Robust deep learning frameworks for acoustic scene and respiratory sound classification | |
| Arora et al. | Automatic Detection and Analysis of Singing Mistakes for Music Pedagogy |
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 |