CN109903754A - Method for voice recognition, equipment and memory devices - Google Patents
Method for voice recognition, equipment and memory devices Download PDFInfo
- Publication number
- CN109903754A CN109903754A CN201711305005.5A CN201711305005A CN109903754A CN 109903754 A CN109903754 A CN 109903754A CN 201711305005 A CN201711305005 A CN 201711305005A CN 109903754 A CN109903754 A CN 109903754A
- Authority
- CN
- China
- Prior art keywords
- semantic
- semanteme
- semantic set
- user
- equipment
- 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
Landscapes
- Machine Translation (AREA)
Abstract
The embodiment of the present disclosure provides a kind of method for voice recognition.This method comprises: receiving the voice input of natural language form from the user；The voice is inputted and carries out semantic word segmentation processing, to obtain the first semantic set of the voice input；Obtain the related with the historical operation of the user second semantic set；Described first semantic set is matched with the described second semantic set；And the historical operation to match in the described second semantic set with the described first semantic set is determined as operational order corresponding with voice input.The embodiment of the present disclosure additionally provides the equipment and storage equipment for speech recognition.
Description
Technical field
This disclosure relates to a kind of method for voice recognition, equipment and memory devices.
Background technique
Interactive voice (natural language interaction) has become the more and more important use of every field (for example, smart home field)
Family entrance.It is used to realize equipment manipulation, Scene realization, interactive cooperation, to realize the simple of human-computer interaction and convenience.
In the prior art, common natural language control function is generally by the dedicated speech recognition technology of industry and spy
Fixed language development tool is realized.The identification of its equipment, equipment control instruction and equipment state analysis etc. are based on specific all kinds of
Equipment carries out preparatory off-line training, and in order to reach accurate control, every class equipment will collect information in advance and be analyzed, be instructed
Practice, forms fixed knowledge and strategy solidification in software.However different user may have for same operation it is various
Phonetic order so that such technology can not identify the targeted operation of user speech well.
Therefore, it is necessary to the technologies that one kind can more effectively identify voice.
Summary of the invention
The one aspect of the embodiment of the present disclosure provides a kind of method for voice recognition.Come this method comprises: receiving
From the voice input of the natural language form of user；The voice is inputted and carries out semantic word segmentation processing, to obtain the voice
The semantic set of the first of input；Obtain the related with the historical operation of the user second semantic set；It is semantic by described first
Set is matched with the described second semantic set；And phase will be gathered with first semanteme in the described second semantic set
The historical operation matched is determined as operational order corresponding with voice input.
Optionally, this method may also include that the facility information for obtaining equipment related with the user；If described second
There is no the historical operation to match with the described first semantic set in semanteme set, identified based on the facility information described
Semanteme in first semantic set；And if all languages in the described first semantic set are identified based on the facility information
Justice, the semanteme in the first semanteme set that will identify that are combined, are referred to forming operation corresponding with voice input
It enables.
Optionally, this method, which may also include that be formed, has same or similar contain with the semanteme in the described first semantic set
The semantic third semanteme set of justice, the third semanteme set are related with the historical operation of the user；If based on described
Facility information does not identify all semantemes in the described first semantic set, will in the described first semantic set it is unrecognized out
Semanteme matched with the semanteme in the third semanteme set；If identified all described in the described first semantic set
Unrecognized semanteme out, the semanteme in the first semanteme set that will identify that is combined, to be formed and voice input
Corresponding operational order.
Optionally, this method may also include that if by the described first semantic set it is unrecognized go out semanteme with it is described
Semanteme in third semanteme set matched do not identify in the described first semantic set it is all it is described it is unrecognized go out
Semanteme, and if user previously carried out speech recognition, by the described first semantic set and user's previous user speech recognition phase
In conjunction with to generate, new first is semantic to gather；By the new first semantic set and the described second semantic set and the third
Semanteme set is matched respectively；And if identify all unrecognized languages out in the described first semantic set
Justice, the semanteme in the first semanteme set that will identify that are combined, are referred to forming operation corresponding with voice input
It enables.
Optionally, this method, which may also include that, records the first semanteme set or the first new semantic set every time
It is matched as a result, updating the described second semantic set and/or the third semanteme set according to each matching result.
The embodiment of the present disclosure another aspect provides a kind of equipment for speech recognition.The equipment includes that voice is defeated
Enter unit, information processing and acquiring unit, sets match unit and operational order determination unit.Voice-input unit is for receiving
The voice of natural language form from the user inputs.Information processing and acquiring unit, which are used to input the voice, carries out semanteme
Word segmentation processing with the obtain voice input first semantic set, and obtains related with the historical operation of the user the
Two semantic set.Sets match unit is used to match the described first semantic set with the described second semantic set.Operation
Instruction-determining unit, for the historical operation to match in the described second semantic set with the described first semantic set to be determined as
Operational order corresponding with voice input.
Optionally, the information processing and acquiring unit are also used to acquiring unit and obtain equipment related with the user
Facility information.The sets match unit is also used to be not present in the described second semantic set and the described first semantic set phase
The semanteme in the described first semantic set is identified when matched historical operation based on the facility information.The operational order is true
Order member is also used to will identify that when identifying all semantemes in the described first semantic set based on the facility information
Semanteme in first semantic set is combined, to form operational order corresponding with voice input.
Optionally, the information processing and acquiring unit are also used to be formed has with the semanteme in the described first semantic set
The semantic third semanteme set of same or similar meaning, the third semanteme set are related with the historical operation of the user.
The sets match unit is also used to do not identifying all languages in the described first semantic set based on the facility information
When adopted, semanteme out unrecognized in the described first semantic set is matched with the semanteme in the third semanteme set.
The operational order determination unit is also used to all unrecognized semantemes out in identifying the described first semantic set
When, the semanteme in the first semanteme set that will identify that is combined, and is referred to forming operation corresponding with voice input
It enables.
Optionally, the information processing and acquiring unit be also used to by the described first semantic set it is unrecognized go out
The semantic semanteme with the third semanteme set matched do not identify in the described first semantic set it is all it is described not
When the semantic and described user being identified previously had carried out speech recognition, by the described first semantic set and user's previous user
Speech recognition combines to generate the semantic set of new first.The sets match unit is also used to the first new semanteme
Set is matched respectively with the described second semantic set and the third semanteme set.The operational order determination unit is also used
When all unrecognized semantemes out in identifying the described first semantic set, the semantic set of will identify that first
In semanteme be combined, to form corresponding with voice input operational order.
Optionally, which may also include data record and updating unit, for recording the described first semantic set or institute
It is matched every time as a result, and updating the described second semantic set according to each matching result to state the semantic set of new first
And/or the third semanteme set.
The embodiment of the present disclosure another aspect provides a kind of equipment for speech recognition.The equipment includes memory
And processor.Memory is for storing executable instruction.Processor is for executing the executable instruction stored in memory, to hold
The row above method.
The embodiment of the present disclosure another aspect provides one kind to carry the memory devices of computer program thereon, when
When executing the computer program by processor, the computer program makes the processor execute the above method.
Detailed description of the invention
For a more complete understanding of the present invention and its advantage, referring now to being described below in conjunction with attached drawing, in which:
Fig. 1 diagrammatically illustrates the outline flowchart of the method for voice recognition according to the embodiment of the present disclosure；
Fig. 2 shows the simplified block diagrams according to the equipment for speech recognition of the embodiment of the present disclosure；
Fig. 3 shows the flow chart of a specific implementation of the method for voice recognition according to the embodiment of the present disclosure；
Fig. 4 shows a signal of the cosine similarity algorithm according to the embodiment of the present disclosure；And
Fig. 5 diagrammatically illustrates the brief block diagram of the electronic equipment according to the embodiment of the present disclosure.
Specific embodiment
Hereinafter, will be described with reference to the accompanying drawings embodiment of the disclosure.However, it should be understood that these descriptions are only exemplary
, and it is not intended to limit the scope of the present disclosure.In addition, in the following description, descriptions of well-known structures and technologies are omitted, with
Avoid unnecessarily obscuring the concept of the disclosure.
Term as used herein is not intended to limit the disclosure just for the sake of description specific embodiment.Used here as
Word " one ", " one (kind) " and "the" etc. also should include " multiple ", " a variety of " the meaning, unless in addition context clearly refers to
Out.In addition, the terms "include", "comprise" as used herein etc. show the presence of the feature, step, operation and/or component,
But it is not excluded that in the presence of or add other one or more features, step, operation or component.
There are all terms (including technical and scientific term) as used herein those skilled in the art to be generally understood
Meaning, unless otherwise defined.It should be noted that term used herein should be interpreted that with consistent with the context of this specification
Meaning, without that should be explained with idealization or excessively mechanical mode.
Shown in the drawings of some block diagrams and/or flow chart.It should be understood that some sides in block diagram and/or flow chart
Frame or combinations thereof can be realized by computer program instructions.These computer program instructions can be supplied to general purpose computer,
The processor of special purpose computer or other programmable data processing units, so that these instructions are when executed by this processor can be with
Creation is for realizing function/operation device illustrated in these block diagrams and/or flow chart.
Therefore, the technology of the disclosure can be realized in the form of hardware and/or software (including firmware, microcode etc.).Separately
Outside, the technology of the disclosure can take the form of the computer program product on the computer-readable medium for being stored with instruction, should
Computer program product uses for instruction execution system or instruction execution system is combined to use.In the context of the disclosure
In, computer-readable medium, which can be, can include, store, transmitting, propagating or transmitting the arbitrary medium of instruction.For example, calculating
Machine readable medium can include but is not limited to electricity, magnetic, optical, electromagnetic, infrared or semiconductor system, device, device or propagation medium.
The specific example of computer-readable medium includes: magnetic memory apparatus, such as tape or hard disk (HDD)；Light storage device, such as CD
(CD-ROM)；Memory, such as random access memory (RAM) or flash memory；And/or wire/wireless communication link.
Definition for smart home device generally all includes product type name, attribute-name, attribute value and related description
Deng.
For example, there is the intelligent attributes (part) of a air-conditioning to be defined as follows:
Device type name: ABC intelligent air condition
Device attribute:
The device attribute list of 1 certain air-conditioning of table
Sometimes user can also name equipment self, for example be " parlor air-conditioning ", " bedroom air-conditioning " etc..
Inventor's discovery, industry is for same class equipment at present, because not seeking unity of standard, can not form unification
Manipulation instruction, therefore certain one kind instruction can not be general.It is switch on a certain lamp by taking switch as an example, is inserted in a certain kind
It may be onoff on seat, be switch on a certain air-conditioning, and may be power for another air-conditioning.In addition, right
The same movement, the description of different user are also not quite similar, such as can have power supply, starting etc. for " switch " movement
Different sayings, even if all users are only concerned meaning as " switch ".
The general control of general intelligence household, be all by customization send the unique manipulation instruction of various kinds of equipment to equipment into
Row control, reading etc., such as transmission<switch, off>instruction close power supply to air-conditioning.
Existing voice controls (natural language control) and carries out equipment manipulation, finally can all be converted to above-mentioned general control.
In order to reach accurate control, every class equipment will collect information in advance and be analyzed, be trained, and form fixed knowledge and strategy
Solidification is in software.Such as " opening air-conditioning ", the prior art trains such air-conditioning in advance, then can be according to air-conditioning category
Switch is " switch ", so that analysis is " setting on for the switch of air-conditioning ".However if encounter a new equipment " soya-bean milk
Machine ", the attribute of switch are " power ".Then to instruction " opening intelligent soy milk grinder ", traditional approach will be unable to identify correct finger
It enables, or is also identified as " setting on for the switch of intelligent soy milk grinder ", and actually intelligent soy milk grinder does not have switch to refer to
It enables, only power instruction, to be unable to reach the purpose of controlling equipment.
To solve the above problems, the embodiment of the present disclosure provides following technical scheme.
Fig. 1 diagrammatically illustrates the outline flowchart of the method for voice recognition according to the embodiment of the present disclosure.
As shown in Figure 1, this method includes operation S110, the voice input of natural language form from the user is received.
In operation s 120, which is inputted and carries out semantic word segmentation processing, with the obtain voice input first semantic collection
It closes.
In operation S130, the related with the historical operation of user second semantic set is obtained.
In operation S140, the first semantic set is matched with the second semantic set.
In some instances, cosine similarity matching can be used to execute operation S140.Language in the first semantic set
When justice and the similarity of the part of semantic in the second semantic set are greater than some threshold value (such as, but not limited to, 0.9), it is believed that should
Two semantic sets match.Certainly, it can also be used in this field and execute operation for carrying out any mode of semantic matches
S140, the embodiment of the present disclosure are not limited by specific matching scheme.
In operation S150, the historical operation to be matched in the second semantic set with the first semantic set is determined as and language
Sound inputs corresponding operational order.
In some instances, this method may also include that the facility information of acquisition equipment related with user.If the second language
There is no the historical operations to match with the first semantic set in justice set, are identified in the first semantic set based on facility information
Semanteme.If identifying all semantemes in the first semantic set, the semantic set of will identify that first based on facility information
In semanteme be combined, to form corresponding with voice input operational order.
For example, the facility information of equipment may include implementor name or mark, device attribute, attribute type, attribute value etc..Work as example
Such as found in facility information with include in the first semantic set implementor name or the identical implementor name of mark or mark when, can be
It is found in the attributes relevant information such as device attribute, attribute type, attribute value and removes implementor name in the first semantic set or identify it
The matching of other outer voices, and so as to form operational order.
In some instances, this method may also include that be formed with the semanteme in the first semantic set with same or similar
The semantic third semanteme set of meaning, third semanteme set are related with the historical operation of user.If do not had based on facility information
There are all semantemes identified in the first semantic set, by semanteme and third semanteme collection out unrecognized in the first semantic set
Semanteme in conjunction is matched.If identify in the first semantic set it is all it is unrecognized go out semantemes, will identify that the
Semanteme in one semantic set is combined, to form operational order corresponding with voice input.
In some instances, this method may also include that if by the first semantic set it is unrecognized go out semanteme and the
Semanteme in three semantic set, which is matched, does not identify all unrecognized semantemes out in the first semantic set, and if
User previously carried out speech recognition, the first semantic set was combined with user's previous user speech recognition to generate new the
One semantic set.New first semantic set is matched respectively with the second semantic set and third semanteme set.If known
Not Chu in the first semantic set it is all it is unrecognized go out semantemes, the semanteme in the semantic set of will identify that first carries out group
It closes, to form operational order corresponding with voice input.
The previous speech recognition of user described herein may include it is last or on arrive used in speech recognition several times
The first semantic set and/or facility information.
In some instances, this method may also include that the semantic set of record first or the new first semantic each of set
It is matching as a result, updating the second semantic set and/or third semanteme set according to each matching result.
No matter whether operation corresponding to user speech is finally identified, all can record the speech recognition as a result, and will
The result or part of it (for example, the semantic component wherein identified) update the semantic collection of above-mentioned second as historical information
Conjunction and/or third semanteme set.
Fig. 2 shows the simplified block diagrams according to the equipment for speech recognition of the embodiment of the present disclosure.
As shown in Fig. 2, the equipment includes voice-input unit 210, information processing and acquiring unit 220, sets match list
Member 230 and operational order determination unit 240.
Voice-input unit 210 is used to receive the voice input of natural language form from the user.
Information processing and acquiring unit 220, which are used to input voice, carries out semantic word segmentation processing, to obtain voice input
First semantic set, and obtain the related with the historical operation of user second semantic set.
Sets match unit 230 is used to match the first semantic set with the second semantic set.
In some instances, cosine similarity matching can be used to execute operation S140.Language in the first semantic set
When justice and the similarity of the part of semantic in the second semantic set are greater than some threshold value (such as, but not limited to, 0.9), it is believed that should
Two semantic sets match.Certainly, it can also be used in this field and execute operation for carrying out any mode of semantic matches
S140, the embodiment of the present disclosure are not limited by specific matching scheme.
Operational order determination unit 240 is used for the historical operation that will be matched in the second semantic set with the first semantic set
It is determined as operational order corresponding with voice input.
In some instances, information processing and acquiring unit 220 are also used to acquiring unit and obtain equipment related with user
Facility information.Sets match unit 230 is also used in the second semantic set that there is no match with the first semantic set
The semanteme in the first semantic set is identified when historical operation based on facility information.Operational order determination unit 240 is also used to
When identifying all semantic in the first semantic set based on facility information, in the semantic set of will identify that first it is semantic into
Row combination, to form operational order corresponding with voice input.
For example, the facility information of equipment may include implementor name or mark, device attribute, attribute type, attribute value etc..Work as example
Such as found in facility information with include in the first semantic set implementor name or the identical implementor name of mark or mark when, can be
It is found in the attributes relevant information such as device attribute, attribute type, attribute value and removes implementor name in the first semantic set or identify it
The matching of other outer voices, and so as to form operational order.
In some instances, information processing and acquiring unit 220 are also used to be formed and the semantic tool in the first semantic set
There is the semantic third semanteme set of same or similar meaning, third semanteme set is related with the historical operation of user.Set
It is also used to unit 230 when not identifying all semantemes in the first semantic set based on facility information, by the first semanteme
Unrecognized semanteme out is matched with the semanteme in third semanteme set in set.Operational order determination unit 240 is also used
Semanteme when all unrecognized semantemes out in identifying the first semantic set, in the semantic set of will identify that first
It is combined, to form operational order corresponding with voice input.
In some instances, information processing and acquiring unit 220 be also used to by the first semantic set it is unrecognized go out
Semanteme matched with the semanteme in third semanteme set do not identify in the first semantic set it is all it is unrecognized out
When semantic and user had previously carried out speech recognition, the first semantic set is combined into next life with user's previous user speech recognition
The semantic set of the first of Cheng Xin.Sets match unit 230 is also used to the new first semantic set and the second semantic set and the
Three semantic set are matched respectively.Operational order determination unit 240 be also used in identifying the first semantic set it is all not
When the semanteme being identified, the semanteme in the semantic set of will identify that first is combined, opposite with voice input to be formed
The operational order answered.
The previous speech recognition of user described herein may include it is last or on arrive used in speech recognition several times
The first semantic set and/or facility information.
In some instances, which may also include data record and updating unit 250, for recording the first semantic set
Or new first semantic set it is matched every time as a result, simultaneously updated according to each matching result the second semantic set and/or
Third semanteme set.
No matter whether operation corresponding to user speech is finally identified, all can record the speech recognition as a result, and will
The result or part of it (for example, the semantic component wherein identified) update the semantic collection of above-mentioned second as historical information
Conjunction and/or third semanteme set.
The technical solution of the embodiment of the present disclosure is retouched above by method shown in FIG. 1 and equipment shown in Fig. 2
It states.The technical solution according to the embodiment of the present disclosure will be described in detail by a specific example below.It should be noted that
The technical solution of the embodiment of the present disclosure is not limited to the specific example, but also may include the example is made fall in guarantor of the present invention
Protect the various modifications in range.
Fig. 3 shows the flow chart of a specific implementation of the method for voice recognition according to the embodiment of the present disclosure.
The input s of the natural language form of user is received first.For the input, can carry out the pretreatment of operation S310 with
And at least one identifying processing in S320-S350.
It will be the skill that illustrates the embodiment of the present disclosure for smart home (especially air-conditioning) in following specific example
Art scheme.It should be appreciated, however, that the technical solution of the embodiment of the present disclosure is similarly applied to carry out using natural-sounding
In any scene of control, and it is not limited to air-conditioning scene.
Natural language pretreatment is executed in operation s 310.
Semantic word segmentation processing, basis of formation semanteme set A can be carried out to original statement s in this operation.Semanteme participle
Processing can be held by using natural language participle technique HanLP (a kind of participle and processing system of the natural language of open source)
Row, but not limited to this, other existing or exploitation in the future any semantic participle processing method/works in this field also can be used
Tool is to execute the operation.
The synonym and near synonym of word in set A are found from predefined smart home thesaurus/near synonym library P,
Together with the original element of set A, it is formed and extends semantic set E1, the nearly justice of semanteme analyzed according to user's history manipulation result
Dictionary U finds word near synonym and the original set of A in set A and together, forms the semantic collection E2 of extension.
Extending being formed for semantic set E1, E2 can be immediately performed later with the semantic word segmentation processing to original statement s, can also
Need using extension semanteme set E1, E2 whenever execution.The execution opportunity of the operation is not limited to operation S310
In.
Where can also obtaining the semantic participle set S that user often manipulates record h and history manipulation sentence with history
Superset conjunction ST=<h1, S1>,<h2, S2>....
By taking air-conditioning as an example, for example, P:{ { open, open, opening, switching }, { switch, power supply } ... }
Illustrate: in such as first set element: opening, open, open, switch as near synonym, other set elements are therewith
It is similar.
U:{ { air-conditioning, parlor air-conditioning }, { temperature, temperature setting } }
ST:{<opening parlor air-conditioning, { opening, parlor air-conditioning }>}
Using air-conditioning as example, it is assumed that air-conditioning is named as parlor air-conditioning, and user successively says following four sentence:
2 original statement of table and its corresponding semantic set
After carrying out semantic word segmentation processing in operation s 310, user is executed in operation S320 and is accustomed to language identification.
Cosine similarity calculation method can be used in this operation.It is of course also possible, as described before, any similarity operator can be used
Method, and it is not limited to cosine similarity algorithm.
To keep the technical solution of the embodiment of the present disclosure clearer, simple Jie first is done to cosine similarity algorithm now
It continues.Fig. 4 shows a signal of the cosine similarity algorithm according to the embodiment of the present disclosure.The algorithm is by calculating between vector
The cos value of angle, judges the similarity of two vectors.
As shown in figure 4, a, b are two vectors, then cosine similarity both are as follows:
By taking following two sentence A and B as an example, wherein word occur number:
Sentence A: I sees/TV at/liking/,/does not like/seeing/film.
Sentence B: whether I/or not do not like/see/TV, also/or not does not like/sees/film.
Then each word frequency of occurrence is as follows:
Sentence A: I 1, likes 2, sees 2, TV 1, film 1, and not 1, also 0.
Sentence B: I 1, likes 2, sees 2, TV 1, film 1, and not 2, also 1.
The then cosine similarity of two sentences are as follows:
Using above-mentioned cosine similarity calculation method, to each of ST history sentence h and participle set S, by h and language
1 (A) of sentence is split as individual character, and the similarity calculation of the history sentence h " opening parlor air-conditioning " in sentence 1 and ST is as follows:
The word frequency of occurrence of sentence 1: it makes a call to 1 sky 1 that 1 opens 1 objective 1 Room 1 and adjusts 1
History sentence " opens parlor air-conditioning ": making a call to 0 sky 1 that 1 opens 1 objective 1 Room 1 and adjusts 1
Therefore cosine similarity are as follows: (1*1+1*1+1*1+1*1+1*0+1*1+1*1)/sqrt (1+1+1+1+0+1+1) *
Sqrt (1+1+1+1+1+1+1)=0.926
If similarity indicates two if being greater than 0.8 using 0.8 as two sentences whether similar judgment threshold is judged
Sentence is similar.Then in the examples described above, value is greater than 0.8, declarative statement 1 and history sentence " opening parlor air-conditioning " very phase
Seemingly, therefore with the operation of history sentence be it is the same, the instruction of the history sentence can be used:<parlor air-conditioning, switch, open>behaviour
Make to carry out equipment manipulation.
It should be noted that the specific structure and content of above-mentioned set S, A, U and ST and dictionary P are only to illustrate this
The example for disclosing embodiment and providing, should not serve to limiting the scope of the invention, also should not serve to equivalent
The first semantic set described in the scheme that Fig. 1 and Fig. 2 are schematically shown, the second semantic set, third semanteme set etc..According to
Concrete implementation (for example, specific participle technique etc.), in other realizations, set S, A, U and ST and dictionary P can also
By with different structure and in the form of.
Assuming that meeting cosine similarity without history sentence for above-mentioned sentence 2,3,4.
It executes and accurately identifies in operation 330.
It needs to carry out the facility information of the equipment of original statement s or its underlying semantics set A and user in this operation
Matching.The facility information may include user possessed implementor name, device product name, equipment remarks, all devices attribute-name,
Attribute type, attribute value range etc..
In this operation, the implementor name or device product for the equipment that the name set of words in A can be possessed with user respectively
Name or device identification or equipment remarks carry out accurate text matches.Whether if mismatching will judge in s includes the implementor name
Or device product name or device identification or equipment remarks, the recognition failures if not including, it can carry out fuzzy in operation S340
Identification.If comprising, then it represents that text matches.
After equipment successful match, remove matched noun from the name set of words of A, and by remaining noun with set
Standby attribute-name is matched.If the verb set of A matched with device attribute without matched attribute, if still do not had
There is matched attribute, then can carry out the fuzzy diagnosis in operation S340.
If device attribute successful match can further judge attribute type, found from the noun of A or number set
Meet the entry of device attribute value range.If successful match, system identification success is accurately identified, is otherwise operated
Fuzzy diagnosis in S340.
For example, including " parlor air-conditioning " in sentence 2, therefore recognizing equipment is " parlor air-conditioning ", and then fixed by air-conditioning
Justice recognizes specific object " wind speed " and its value " low grade ", to have found specific operational order by accurate match:
<parlor air-conditioning, wind speed, low-grade>, manipulation is completed by instruction.
Assuming that sentence 3 and 4 cannot find exact instruction.
Fuzzy diagnosis is executed in operation S340.
The operation is using the result LR in accurately identifying, the semantic set E2 of extension；Equipment di (0 <=i under user
<=number of user equipment) facility information.Wherein, can to implementor name, device product name, equipment remarks carry out word segmentation processing with
Form set Di.All set Di constitute set DA={ Di } (0 <=i <=number of user equipment).
If be handled as follows: without including facility information to DA={ Di } (0 <=i <=user device quantity in LR
Amount) each of set Di, search near synonym in user's history manipulation result semanteme near synonym library U, merge with former Di, shape
The set DDi (0 <=i <=number of user equipment) of Cheng Xin.If DDi is the subclass of E2, matching unit di is successful, into
Enter second step, enters next subsystem if all DDi traversal is done without matching.
If having included facility information or above equipment di successful match in LR, first removed from the noun in E2
Element in the DDi set of matching unit di, obtains EDDi.Title and enumerated value description to device attribute pj is (non-to enumerate
Value then only includes title), form set Fj.It searches near synonym library P and U and Fj to merge, obtaining set FFj, (0 <=j <=is set
Standby attribute number).If FFj is the subclass of EDDi, all matched FFj and attribute newly gather composition one,
I.e. SD=<pj,FFj>(0 <=j <=equipment attribute number), if SD be sky, can carry out at context identification
Manage S350.Otherwise, pair that element is most in the FFj of the inside SD is found out<pj,FFj>, as fuzzy diagnosis as a result, and carrying out equipment
Manipulation.If in recognition result, attribute-name can only be matched to, be described without attribute value, then illustrate it is not enumeration type, to the category
Property data type is judged (data type is generally integer, one kind of floating type), specific number is found, as the attribute
Value manipulated.
Sentence 3: air-conditioner temperature is set as 25 degree, it is assumed that does not all find facility information in operation S320 and S330.
Therefore it is handled using the first step, obtains DDi={ air-conditioning, parlor air-conditioning }, it is possible to determine that DDi is the son of E2
Collection, therefore, recognizes equipment " parlor air-conditioning ".Second step, obtain EDDi=temperature, if being, 25, degree, temperature setting }；It is right
In Fj (j=3), corresponding pj is " temperature setting " this attribute) expanded set FFj={ temperature setting, temperature }, FFj is
The subclass of EDDi is matched to an attribute " temperature setting ", equally can be determined that other pj (j<>3) are unsatisfactory for set and determine
Condition.Therefore, " temperature setting " this attribute is finally matched, is integer according to its type, finds 25 for attribute value, to find
Specific control instruction:<parlor air-conditioning, temperature setting, 25>.
Equipment is not found in sentence 4.
In operation S350, context identification is executed
This can be used to input semantic set A in this operation, it is successful to identify if last user identifies successfully
Device name and attribute-name form set A1, and last user inputs semantic set AL.
If A1 is not empty, new set AA=A+A1；Otherwise the set of A and AL is merged, forms new semantic collection AA；
Input by new AA set as substitution A, successively uses " accurate system ", " fuzzy diagnosis ", " context
The processing logic of identification " completes identification as long as one of identify manipulation logic, carry out practical manipulation；If this
A little systems all do not identify manipulation logic, then fail the intention of correct understanding user to judgement system in operation S370, and
Feed back to user.
Such as sentence 4:A1={ parlor air-conditioning, temperature setting }, therefore AA=parlor air-conditioning, and temperature setting, degree, too,
It is cold, 28 }.Using AA as new A, after carrying out second step operation, use can be recognized inside " logic of fuzzy recognition system "
The recognition result at family:<parlor air-conditioning, temperature setting, 28>, so as to complete to manipulate.
No matter whether aforesaid operations S320-S350 identifies intention, in operation S360, in actually manipulation or returns
Before, analysis result feedback is all arrived into the system, which carries out the training of natural language corpus by technologies such as machine learning, into
One-step optimization basic algorithm and near synonym library P and U.
Although being not meant to that aforesaid operations are needed according to the suitable of number it should be noted that aforesaid operations are numbered
Sequence executes, for example, accurately identifying in operation S330 can also execute after operate S310 and before operating S320, or
It is executed with other any feasible sequences.Therefore, unless explicitly stated otherwise, otherwise operation described herein is not required according to volume
Number sequence execute.
Above-described embodiment of the disclosure can complete natural language and manipulate intelligence in real time dynamically according to the various information that can be obtained
Energy household, solving existing natural language control system analytical equipment information and user's corpus could accurately must manipulate and set in advance
Standby problem, the saying new for the equipment added in real time and user can also complete the purpose of identification.
Not only by facility information, also realize that natural language is grasped in combination with user information, user's natural language habit
Smart home is controlled, natural language manipulation smart home efficiency and accuracy are greatly improved.
Fig. 5 diagrammatically illustrates the block diagram of equipment according to an embodiment of the present disclosure.Equipment shown in Fig. 5 is only one
Example, should not function to the embodiment of the present disclosure and use scope bring any restrictions.
As shown in figure 5, equipment 500 includes central processing unit (CPU) 501 according to this embodiment, it can be according to depositing
Storage is loaded into random access storage device (RAM) 503 in the program in read-only memory (ROM) 502 or from storage section 508
Program and execute various movements appropriate and processing.In RAM 503, also it is stored with equipment 500 and operates required various journeys
Sequence and data.CPU 501, ROM 502 and RAM 503 are connected with each other by bus 504.Input/output (I/O) interface 505
It is also connected to bus 504.
Equipment 500 can also include be connected to I/O interface 505 with one or more in lower component: including keyboard or
The importation 506 of mouse etc.；Including cathode-ray tube (CRT) or liquid crystal display (LCD) etc. and loudspeaker etc.
Output par, c 507；Storage section 508 including hard disk etc.；And the network interface including LAN card or modem etc.
The communications portion 509 of card.Communications portion 509 executes communication process via the network of such as internet.Driver 510 is also according to need
It is connected to I/O interface 505.Detachable media 511, such as disk, CD, magneto-optic disk or semiconductor memory etc., according to
It needs to be mounted on driver 510, in order to be mounted into storage section as needed from the computer program read thereon
508。
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer-readable medium
On computer program, which includes the program code for method shown in execution flow chart.In such reality
It applies in example, which can be downloaded and installed from network by communications portion 509, and/or from detachable media
511 are mounted.When the computer program is executed by central processing unit (CPU) 501, execute in the equipment of the embodiment of the present disclosure
The above-mentioned function of limiting.
It should be noted that computer-readable medium shown in the disclosure can be computer-readable signal media or meter
Calculation machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example can be --- but not
Be limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.It calculates
The more specific example of machine readable storage medium storing program for executing can include but is not limited to: have the electrical connection, portable of one or more conducting wires
Formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable programmable read only memory
(EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device or
Above-mentioned any appropriate combination.In the disclosure, computer readable storage medium can be it is any include or storage program
Tangible medium, the program can be commanded execution system, device or device use or in connection.And in the disclosure
In, computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal, wherein
Carry computer-readable program code.The data-signal of this propagation can take various forms, including but not limited to electric
Magnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be computer-readable and deposit
Any computer-readable medium other than storage media, the computer-readable medium can send, propagate or transmit for by referring to
Enable execution system, device or device use or program in connection.The program for including on computer-readable medium
Code can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable or RF etc. or above-mentioned times
The suitable combination of meaning.
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the disclosure, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, section or code of table, a part of above-mentioned module, section or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it to infuse
It anticipates, the combination of each box in block diagram or flow chart and the box in block diagram or flow chart can be used and execute regulation
The dedicated hardware based systems of functions or operations realize, or can be using a combination of dedicated hardware and computer instructions
To realize.
Such as field programmable gate can also be used according to the method, apparatus of each embodiment of the disclosure, unit and/or module
Array (FPGA), programmable logic array (PLA), system on chip, the system on substrate, the system in encapsulation, dedicated integrated electricity
Road (ASIC) can be for carrying out the hardware such as any other rational method that is integrated or encapsulating or firmware in fact to circuit
It is existing, or realized with software, the appropriately combined of three kinds of implementations of hardware and firmware.The system may include storage equipment,
To realize storage as described above.When realizing in such ways, used software, hardware and/or firmware be programmed or
It is designed as executing the corresponding above method, step and/or the function according to the disclosure.Those skilled in the art can be according to practical need
Come suitably by one or more of these systems and module, or in which a part or multiple portions use it is different upper
Implementation is stated to realize.These implementations each fall within protection scope of the present invention.
As the skilled person will appreciate, for any and all purpose, such as written theory is being provided
The aspect of bright book, all ranges disclosed herein are also covered by any and all possible subrange and its son
The combination of range.Any listed range, which can be readily identified into, adequately describes and enables same range
At least it is broken down into same two parts, three parts, four parts, five parts, ten parts, etc..As a non-limiting example,
Each range discussed herein can be easily decomposed into lower one third, middle one third and upper three/
One etc..Such as those skilled in the art it will also be understood that, all languages of " until ", " at least ", " being greater than ", " being less than " etc.
Yan Jun includes stated quantity and is the range for referring to be broken down into subrange as discussed above therewith.Finally,
As the skilled person will appreciate, range includes each individual ingredient.So for example, the group with 1-3 unit
Refer to the group with 1,2 or 3 unit.Similarly, the group with 1-5 unit refers to 1,2,3,4 or 5 unit
Group, etc..
Although the present invention, art technology has shown and described referring to the certain exemplary embodiments of the disclosure
Personnel it should be understood that in the case where the spirit and scope of the present invention limited without departing substantially from the following claims and their equivalents,
A variety of changes in form and details can be carried out to the disclosure.Therefore, the scope of the present invention should not necessarily be limited by above-described embodiment,
But should be not only determined by appended claims, also it is defined by the equivalent of appended claims.
Claims (12)
1. a kind of method for voice recognition, comprising:
Receive the voice input of natural language form from the user；
The voice is inputted and carries out semantic word segmentation processing, to obtain the first semantic set of the voice input；
Obtain the related with the historical operation of the user second semantic set；
Described first semantic set is matched with the described second semantic set；And
The historical operation to match in described second semantic set with the described first semantic set is determined as defeated with the voice
Enter corresponding operational order.
2. according to the method described in claim 1, further include:
Obtain the facility information of equipment related with the user；
If set there is no the historical operation to match with the described first semantic set based on described in the described second semantic set
Standby information identifies the semanteme in the described first semantic set；And
If identifying all semantemes in the described first semantic set based on the facility information, will identify that first is semantic
Semanteme in set is combined, to form operational order corresponding with voice input.
3. according to the method described in claim 2, further include:
The semantic third semanteme set that there is same or similar meaning with the semanteme in the described first semantic set is formed, it is described
Third semanteme set is related with the historical operation of the user；
It is semantic by described first if not identifying all semantemes in the described first semantic set based on the facility information
Unrecognized semanteme out is matched with the semanteme in the third semanteme set in set；
If identifying all unrecognized semantemes out, the semantic collection of will identify that first in the described first semantic set
Semanteme in conjunction is combined, to form operational order corresponding with voice input.
4. according to the method described in claim 3, further include:
If by the described first semantic set it is unrecognized go out semanteme and semantic in the third semanteme set carry out
With do not identify in the described first semantic set it is all it is described it is unrecognized go out semantemes, and if user previously carried out language
Sound identification combines the described first semantic set to generate the semantic set of new first with user's previous user speech recognition；
The first new semantic set is matched respectively with the described second semantic set and the third semanteme set；With
And
If identifying all unrecognized semantemes out, the semantic collection of will identify that first in the described first semantic set
Semanteme in conjunction is combined, to form operational order corresponding with voice input.
5. method according to any of claims 1-4, further includes:
It records the described first semantic set or the first new semantic set is matched as a result, according to each matching knot every time
Fruit updates the described second semantic set and/or the third semanteme set.
6. a kind of equipment for speech recognition, comprising:
Voice-input unit, the voice for receiving natural language form from the user input；
Information processing and acquiring unit carry out semantic word segmentation processing for inputting to the voice, to obtain the voice input
The first semantic set, and obtain related with the historical operation of the user second and semantic gather；
Sets match unit, for matching the described first semantic set with the described second semantic set；And
Operational order determination unit, for grasping the history to match in the described second semantic set with the described first semantic set
Work is determined as operational order corresponding with voice input.
7. equipment according to claim 6, in which:
The information processing and acquiring unit are also used to the facility information that acquiring unit obtains equipment related with the user；
The sets match unit is also used to be not present in the described second semantic set to match with the described first semantic set
Historical operation when semanteme in the described first semantic set is identified based on the facility information；And
The operational order determination unit is also used to identifying the institute in the described first semantic set based on the facility information
When having semanteme, the semanteme in the semantic set of will identify that first is combined, corresponding with voice input to be formed
Operational order.
8. equipment according to claim 7, further includes:
The information processing and acquiring unit are also used to be formed with the semanteme in the described first semantic set with same or similar
The semantic third semanteme set of meaning, the third semanteme set are related with the historical operation of the user；
The sets match unit is also used to do not identifying the institute in the described first semantic set based on the facility information
When having semanteme, by the described first semantic set it is unrecognized go out semanteme and semantic in the third semanteme set carry out
Match；
The operational order determination unit be also used in identifying the described first semantic set it is all it is described it is unrecognized go out
When semantic, the semanteme in the semantic set of will identify that first is combined, to form behaviour corresponding with voice input
It instructs.
9. equipment according to claim 8, further includes:
The information processing and acquiring unit be also used to by the described first semantic set it is unrecognized go out semanteme with it is described
Semanteme in third semanteme set matched do not identify in the described first semantic set it is all it is described it is unrecognized go out
When the semantic and described user had previously carried out speech recognition, by the described first semantic set and user's previous user speech recognition phase
In conjunction with to generate, new first is semantic to gather；
The sets match unit is also used to the new first semantic set and the described second semantic set and the third
Semanteme set is matched respectively；And
The operational order determination unit be also used in identifying the described first semantic set it is all it is described it is unrecognized go out
When semantic, the semanteme in the semantic set of will identify that first is combined, to form behaviour corresponding with voice input
It instructs.
10. the equipment according to any one of claim 6-9, further includes:
Data record and updating unit, for recording the described first semantic set or the first new semantic set matching every time
As a result, simultaneously updating the described second semantic set and/or the third semanteme set according to each matching result.
11. a kind of equipment for for speech recognition, comprising:
Memory, for storing executable instruction；And
Processor, for executing the executable instruction stored in memory, to execute according to claim 1 described in any one of -5
Method.
12. one kind carries the memory devices of computer program, when executing the computer program by processor, institute thereon
Stating computer program makes the processor execute method according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711305005.5A CN109903754B (en) | 2017-12-08 | 2017-12-08 | Method, device and memory device for speech recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711305005.5A CN109903754B (en) | 2017-12-08 | 2017-12-08 | Method, device and memory device for speech recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109903754A true CN109903754A (en) | 2019-06-18 |
| CN109903754B CN109903754B (en) | 2022-04-26 |
Family
ID=66941737
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711305005.5A Active CN109903754B (en) | 2017-12-08 | 2017-12-08 | Method, device and memory device for speech recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109903754B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110767219A (en) * | 2019-09-17 | 2020-02-07 | 中国第一汽车股份有限公司 | Semantic updating method, device, server and storage medium |
| CN110838292A (en) * | 2019-09-29 | 2020-02-25 | 广东美的白色家电技术创新中心有限公司 | Voice interaction method, electronic equipment and computer storage medium |
Citations (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07334504A (en) * | 1994-06-08 | 1995-12-22 | Meidensha Corp | Japanese processing system |
| JPH08146990A (en) * | 1994-11-24 | 1996-06-07 | Matsushita Electric Ind Co Ltd | Natural language interaction device |
| US20060184353A1 (en) * | 2000-07-20 | 2006-08-17 | Microsoft Corporation | Ranking Parser for a Natural Language Processing System |
| US20090076795A1 (en) * | 2007-09-18 | 2009-03-19 | Srinivas Bangalore | System And Method Of Generating Responses To Text-Based Messages |
| US20130166280A1 (en) * | 2011-12-21 | 2013-06-27 | Nuance Communications, Inc. | Concept Search and Semantic Annotation for Mobile Messaging |
| CN104290097A (en) * | 2014-08-19 | 2015-01-21 | 白劲实 | Learning type intelligent home social contact robot system and method |
| US20150046168A1 (en) * | 2013-08-06 | 2015-02-12 | Nuance Communications, Inc. | Method and Apparatus for a Multi I/O Modality Language Independent User-Interaction Platform |
| CN104951456A (en) * | 2014-03-26 | 2015-09-30 | 上海智臻网络科技有限公司 | Method, device and equipment used for obtaining answer information |
| CN105389400A (en) * | 2015-12-24 | 2016-03-09 | Tcl集团股份有限公司 | Speech interaction method and device |
| CN106445924A (en) * | 2016-11-14 | 2017-02-22 | Tcl集团股份有限公司 | Method and system for controlling intelligent equipment on basis of semantic server |
| CN106503239A (en) * | 2016-11-07 | 2017-03-15 | 上海智臻智能网络科技股份有限公司 | A kind of method and apparatus of legal information inquiry |
| CN106601248A (en) * | 2017-01-20 | 2017-04-26 | 浙江小尤鱼智能技术有限公司 | Smart home system based on distributed voice control |
| US20170140750A1 (en) * | 2015-11-17 | 2017-05-18 | Le Holdings (Beijing) Co., Ltd. | Method and device for speech recognition |
| CN106792060A (en) * | 2017-01-11 | 2017-05-31 | 微鲸科技有限公司 | The control method and device of external equipment |
| CN106847269A (en) * | 2017-01-20 | 2017-06-13 | 浙江小尤鱼智能技术有限公司 | The sound control method and device of a kind of intelligent domestic system |
| CN106921544A (en) * | 2016-06-20 | 2017-07-04 | 广州零号软件科技有限公司 | The Intelligent household voice control system of default interactive voice order dictionary |
| CN106933561A (en) * | 2015-12-31 | 2017-07-07 | 北京搜狗科技发展有限公司 | Pronunciation inputting method and terminal device |
| US20170236510A1 (en) * | 2016-02-17 | 2017-08-17 | Honda Motor Co., Ltd. | Voice processing device |
| CN107146622A (en) * | 2017-06-16 | 2017-09-08 | 合肥美的智能科技有限公司 | Refrigerator, voice interactive system, method, computer equipment, readable storage medium storing program for executing |
| CN107170449A (en) * | 2017-06-14 | 2017-09-15 | 上海雍敏信息科技有限公司 | Intelligent domestic system and its control method |
-
2017
- 2017-12-08 CN CN201711305005.5A patent/CN109903754B/en active Active
Patent Citations (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07334504A (en) * | 1994-06-08 | 1995-12-22 | Meidensha Corp | Japanese processing system |
| JPH08146990A (en) * | 1994-11-24 | 1996-06-07 | Matsushita Electric Ind Co Ltd | Natural language interaction device |
| US20060184353A1 (en) * | 2000-07-20 | 2006-08-17 | Microsoft Corporation | Ranking Parser for a Natural Language Processing System |
| US20090076795A1 (en) * | 2007-09-18 | 2009-03-19 | Srinivas Bangalore | System And Method Of Generating Responses To Text-Based Messages |
| US20130166280A1 (en) * | 2011-12-21 | 2013-06-27 | Nuance Communications, Inc. | Concept Search and Semantic Annotation for Mobile Messaging |
| US20150046168A1 (en) * | 2013-08-06 | 2015-02-12 | Nuance Communications, Inc. | Method and Apparatus for a Multi I/O Modality Language Independent User-Interaction Platform |
| CN104951456A (en) * | 2014-03-26 | 2015-09-30 | 上海智臻网络科技有限公司 | Method, device and equipment used for obtaining answer information |
| CN104290097A (en) * | 2014-08-19 | 2015-01-21 | 白劲实 | Learning type intelligent home social contact robot system and method |
| US20170140750A1 (en) * | 2015-11-17 | 2017-05-18 | Le Holdings (Beijing) Co., Ltd. | Method and device for speech recognition |
| CN105389400A (en) * | 2015-12-24 | 2016-03-09 | Tcl集团股份有限公司 | Speech interaction method and device |
| CN106933561A (en) * | 2015-12-31 | 2017-07-07 | 北京搜狗科技发展有限公司 | Pronunciation inputting method and terminal device |
| US20170236510A1 (en) * | 2016-02-17 | 2017-08-17 | Honda Motor Co., Ltd. | Voice processing device |
| CN106921544A (en) * | 2016-06-20 | 2017-07-04 | 广州零号软件科技有限公司 | The Intelligent household voice control system of default interactive voice order dictionary |
| CN106503239A (en) * | 2016-11-07 | 2017-03-15 | 上海智臻智能网络科技股份有限公司 | A kind of method and apparatus of legal information inquiry |
| CN106445924A (en) * | 2016-11-14 | 2017-02-22 | Tcl集团股份有限公司 | Method and system for controlling intelligent equipment on basis of semantic server |
| CN106792060A (en) * | 2017-01-11 | 2017-05-31 | 微鲸科技有限公司 | The control method and device of external equipment |
| CN106847269A (en) * | 2017-01-20 | 2017-06-13 | 浙江小尤鱼智能技术有限公司 | The sound control method and device of a kind of intelligent domestic system |
| CN106601248A (en) * | 2017-01-20 | 2017-04-26 | 浙江小尤鱼智能技术有限公司 | Smart home system based on distributed voice control |
| CN107170449A (en) * | 2017-06-14 | 2017-09-15 | 上海雍敏信息科技有限公司 | Intelligent domestic system and its control method |
| CN107146622A (en) * | 2017-06-16 | 2017-09-08 | 合肥美的智能科技有限公司 | Refrigerator, voice interactive system, method, computer equipment, readable storage medium storing program for executing |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110767219A (en) * | 2019-09-17 | 2020-02-07 | 中国第一汽车股份有限公司 | Semantic updating method, device, server and storage medium |
| CN110767219B (en) * | 2019-09-17 | 2021-12-28 | 中国第一汽车股份有限公司 | Semantic updating method, device, server and storage medium |
| CN110838292A (en) * | 2019-09-29 | 2020-02-25 | 广东美的白色家电技术创新中心有限公司 | Voice interaction method, electronic equipment and computer storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109903754B (en) | 2022-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109036384B (en) | Audio recognition method and device | |
| CN111026842B (en) | Natural language processing method, natural language processing device and intelligent question-answering system | |
| CN107679039A (en) | The method and apparatus being intended to for determining sentence | |
| CN108628947A (en) | A kind of business rule matched processing method, device and processing equipment | |
| CN111783439B (en) | Man-machine interaction dialogue processing method and device, computer equipment and storage medium | |
| CN110489751A (en) | Text similarity computing method and device, storage medium, electronic equipment | |
| CN109165380A (en) | A kind of neural network model training method and device, text label determine method and device | |
| CN105446725B (en) | Method and system for Mode-driven architecture | |
| CN109635094A (en) | Method and apparatus for generating answer | |
| CN108256070A (en) | For generating the method and apparatus of information | |
| CN109976702A (en) | A kind of audio recognition method, device and terminal | |
| CN107506434A (en) | Method and apparatus based on artificial intelligence classification phonetic entry text | |
| CN109299477A (en) | Method and apparatus for generating text header | |
| CN109783624A (en) | Answer generation method, device and the intelligent conversational system in knowledge based library | |
| CN110019742A (en) | Method and apparatus for handling information | |
| CN109858045A (en) | Machine translation method and device | |
| CN108257593A (en) | A kind of audio recognition method, device, electronic equipment and storage medium | |
| CN109885657A (en) | A kind of calculation method of text similarity, device and storage medium | |
| CN108986805A (en) | Method and apparatus for sending information | |
| CN109582954A (en) | Method and apparatus for output information | |
| CN110362800A (en) | Configuration method, device, electronic equipment and the medium of product information | |
| CN110275963A (en) | Method and apparatus for output information | |
| CN109582788A (en) | Comment spam training, recognition methods, device, equipment and readable storage medium storing program for executing | |
| CN108959529A (en) | Determination method, apparatus, equipment and the storage medium of problem answers type | |
| CN110516261A (en) | Resume appraisal procedure, device, electronic equipment and computer storage medium |
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 |