CN106682642A - Multi-language-oriented behavior identification method and multi-language-oriented behavior identification system - Google Patents
Multi-language-oriented behavior identification method and multi-language-oriented behavior identification system Download PDFInfo
- Publication number
- CN106682642A CN106682642A CN201710012063.2A CN201710012063A CN106682642A CN 106682642 A CN106682642 A CN 106682642A CN 201710012063 A CN201710012063 A CN 201710012063A CN 106682642 A CN106682642 A CN 106682642A
- Authority
- CN
- China
- Prior art keywords
- information
- language performance
- language
- user
- characteristic information
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000000605 extraction Methods 0.000 claims abstract description 39
- 230000000694 effects Effects 0.000 claims description 8
- 239000000284 extract Substances 0.000 claims description 2
- 238000013473 artificial intelligence Methods 0.000 abstract description 4
- 230000006399 behavior Effects 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 6
- 238000012552 review Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000005194 fractionation Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000007637 random forest analysis Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/70—Multimodal biometrics, e.g. combining information from different biometric modalities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/40—Document-oriented image-based pattern recognition
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Computer Interaction (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Acoustics & Sound (AREA)
- Machine Translation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention belongs to the technical field of artificial intelligence dialogue, and provides a multi-language-oriented behavior identification method and a multi-language-oriented behavior identification system. The method comprises the steps of acquiring current information input by a user, user figure information and external information; performing information extraction on the current information for extracting characteristic information; and performing language behavior identification according to the characteristic information, the user figure information and the external information, thereby obtaining a language behavior of the user. The multi-language-oriented behavior identification method and the multi-language-oriented behavior identification system can accurately and effectively identify the language behavior of the user, thereby reducing errir identification rate for the language behavior of the user.
Description
Technical field
The present invention relates to artificial intelligence dialogue technoloyg field, and in particular to much a kind of towards language performance recognition methods and to be
System.
Background technology
Existing artificial intelligence conversational system, generally by the way of hyphenation subordinate sentence, phrase is split into by sentence, is divided
Analysis, obtains extra information, such as topic, mood and keyword.But, sentence will lose language semantically after splitting
Behavior, causes artificial intelligence conversational system to provide appropriate reply.
Also, user language Activity recognition is that text information based on user input, acoustic information predict user's mostly
Language performance.Due to the environmental information where lacking user, and user's portrait information, such as past historical information of user is led
Cause the distortion of the result of prediction.
How precisely, the language performance of user is efficiently identified, user language behavior False Rate is reduced, is art technology
The problem of personnel's urgent need to resolve.
The content of the invention
For defect of the prior art, the present invention provides a kind of many towards language performance recognition methods and system, can
Precisely, the language performance of user is efficiently identified, user language behavior False Rate is reduced.
In a first aspect, the present invention provides a kind of many towards language performance recognition methods, the method includes：
Obtain current information, user portrait information and the external information of user input；
Information extraction is carried out to current information, characteristic information is obtained；
According to characteristic information, user portrait information and external information, language performance identification is carried out, obtain the language row of user
For.
Further, information extraction is carried out to current information, characteristic information is obtained, is specifically included：
Semantics information extraction is carried out to text message, text feature information is obtained；
Sound characteristic extraction is carried out to acoustic information, sound characteristic information is obtained；
Image feature extraction is carried out to image information, image character information is obtained,
Current information includes text message, acoustic information and image information,
Characteristic information includes text feature information, sound characteristic information and image character information；
According to characteristic information, user portrait information and external information, language performance identification is carried out, obtain the language row of user
To specifically include：
According to text feature information, sound characteristic information, image character information, user portrait information and external information, enter
Row language performance is recognized, and record instruction Activity recognition number of times, if language performance is recognized successfully, generates the language row of the user
For, if language performance recognition failures, language performance identification is carried out again, until language performance is recognized successfully, or language performance
Identification number of times reaches preset value.
Further, after text feature information is obtained, before carrying out language performance identification, the method also includes：Root
According to text message and text feature information, review text information.
Based on above-mentioned any many towards language performance recognition methods embodiment, further, language performance identification is being carried out
Before, the method also includes：
Obtain sample text characteristic information, sample audio characteristic information, sample image characteristic information, user's portrait information and
External information；
According to sample text characteristic information, sample audio characteristic information, sample image characteristic information, user portrait information and
External information, carries out language performance identification, obtains specimen discerning language performance；
The similarity of the true language performance of sample and specimen discerning language performance according to pre-acquiring, sets preset value.
Second aspect, the present invention provides a kind of many towards language performance identifying system, and the system includes acquisition of information subsystem
System, information extraction subsystem and language performance recognition subsystem, information acquisition subsystem are used to obtain the current letter of user input
Breath, user portrait information and external information；Information extraction subsystem is used to carry out current information information extraction, obtains feature letter
Breath；Language performance recognition subsystem is used to, according to characteristic information, user portrait information and external information, carry out language performance knowledge
Not, the language performance of user is obtained.
Further, information extraction subsystem specifically for：Semantics information extraction is carried out to text message, text is obtained special
Reference ceases；Sound characteristic extraction is carried out to acoustic information, sound characteristic information is obtained；Image feature is carried out to image information to carry
Take, obtain image character information, current information includes text message, acoustic information and image information, and characteristic information includes text
Characteristic information, sound characteristic information and image character information；
Language performance recognition subsystem specifically for：According to text feature information, sound characteristic information, image feature letter
Breath, user portrait information and external information, carry out language performance identification, and record instruction Activity recognition number of times, if language performance
Recognize successfully, then generate the language performance of the user, if language performance recognition failures, language performance identification is carried out again, directly
Recognized successfully to language performance, or language performance identification number of times reaches preset value.
Further, the present embodiment is more also includes text message amendment subsystem towards language performance identifying system, is used for
According to text message and text feature information, review text information.
Based on above-mentioned arbitrarily more towards language performance identifying system embodiment, further, the system also includes preset value
Subsystem is determined, for obtaining sample text characteristic information, sample audio characteristic information, sample image characteristic information, Yong Huhua
As information and external information；According to sample text characteristic information, sample audio characteristic information, sample image characteristic information, user
Portrait information and external information, carry out language performance identification, obtain specimen discerning language performance；Sample according to pre-acquiring is true
The similarity of language performance and specimen discerning language performance, sets preset value.
As shown from the above technical solution, the present embodiment is more towards language performance recognition methods and system, can be according to user
The many current informations towards language performance of input, carry out feature information extraction, it is to avoid enter line statement fractionation, to be conducive to basis to work as
Preceding information engages in the dialogue reply.Meanwhile, the method combination user portrait information and external information carry out language performance identification, energy
Language performance identification error situation is enough reduced, language performance False Rate is reduced.
Therefore, the present embodiment is more towards language performance recognition methods and system, can precisely, efficiently identify the language of user
Words and deeds are to reduce user language behavior False Rate, are favorably improved communication efficiency.
Brief description of the drawings
In order to illustrate more clearly of the specific embodiment of the invention or technical scheme of the prior art, below will be to specific
The accompanying drawing to be used needed for implementation method or description of the prior art is briefly described.In all of the figs, similar element
Or the general reference by being similar in part is identified.In accompanying drawing, each element or part might not draw according to actual ratio.
Fig. 1 shows that one kind that the embodiment of the present invention is provided is more towards language performance recognition methods flow chart；
Fig. 2 shows a kind of many structural representations towards language performance identifying system that the embodiment of the present invention is provided；
Fig. 3 shows many structural representations towards language performance identifying system of another kind that the embodiment of the present invention is provided
Figure.
Specific embodiment
The embodiment of technical solution of the present invention is described in detail below in conjunction with accompanying drawing.Following examples are only used for
Technical scheme is clearly illustrated, therefore is intended only as example, and protection of the invention can not be limited with this
Scope.
It should be noted that unless otherwise indicated, technical term used in this application or scientific terminology should be this hair
The ordinary meaning that bright one of ordinary skill in the art are understood.
In a first aspect, the embodiment of the present invention provides a kind of many towards language performance recognition methods, with reference to Fig. 1, the method bag
Include：
Step S1, obtains current information, user portrait information and the external information of user input, current information be it is many towards
The information of language performance, here, current information can be text message, acoustic information and image information, user's portrait information is
The information such as custom, preference, the operation note of user, external information is the outside daily letter such as place to use, use time and news
Breath.
Step S2, information extraction is carried out to current information, obtains characteristic information.
Step S3, according to characteristic information, user portrait information and external information, carries out language performance identification, obtains user
Language performance.
As shown from the above technical solution, the present embodiment is more towards language performance recognition methods, can be many according to user input
Towards the current information of language performance, feature information extraction is carried out, it is to avoid enter line statement fractionation, to be conducive to according to current information
Engage in the dialogue reply.Meanwhile, the method combination user portrait information and external information carry out language performance identification, can reduce
Language performance recognizes error situation, reduces language performance False Rate.
Therefore, the present embodiment is more towards language performance recognition methods, can precisely, efficiently identify the language row of user
For, user language behavior False Rate is reduced, it is favorably improved communication efficiency.
Specifically, the present embodiment is more can be processed current information towards language performance recognition methods, and it was realized
Journey is as follows：
Semantics information extraction is carried out to text message, text feature information is obtained；Here, using semantics information extraction side
Method, extracts the meaning of one's words, obtains text feature information.
Sound characteristic extraction is carried out to acoustic information, sound characteristic information is obtained；Here, using sound characteristic extraction machine
System, obtains sound characteristic information.
Image feature extraction is carried out to image information, image feature extracting mechanism is such as used, image character information is obtained, when
Preceding information include text message, acoustic information and image information, characteristic information include text feature information, sound characteristic information and
Image character information；
According to characteristic information, user portrait information and external information, language performance identification is carried out, obtain the language row of user
For when, implementation process is as follows：According to text feature information, sound characteristic information, image character information, user's portrait information and outer
Portion's information, carries out language performance identification, and record instruction Activity recognition number of times, if language performance is recognized successfully, generates the use
The language performance at family, if language performance recognition failures, carries out language performance identification again, until language performance is recognized successfully,
Or language performance identification number of times reaches preset value.
Here, the present embodiment is more can to process many information, such as text message, sound towards language performance recognition methods
Message ceases and image information, according at least one of text feature information, sound characteristic information, image character information three,
And user portrait information and external information are combined, language performance identification is carried out, wherein, according to text feature information, sound characteristic
Information, image character information, user portrait information and external information, carry out language performance identification, obtain the accurate of language performance
Degree highest.Meanwhile, in language performance recognition failures, the method can also carry out repeating identification, to obtain the language row of user
To be conducive to improving the stability and Consumer's Experience of the method.
Also, the present embodiment is more can also to set iterations towards language performance recognition methods in the sample training stage,
I.e. preset value, implements process as follows：
Obtain sample text characteristic information, sample audio characteristic information, sample image characteristic information, user's portrait information and
External information.According to sample text characteristic information, sample audio characteristic information, sample image characteristic information, user's portrait information
And external information, language performance identification is carried out, obtain specimen discerning language performance.The true language performance of sample according to pre-acquiring
With the similarity of specimen discerning language performance, preset value is set.
Here, the method sets iterations, i.e. preset value according to the sample of training, with Optimal Parameters, it is favorably improved
The accuracy and treatment effeciency of language performance identification.
Specifically, after text feature information is obtained, before carrying out language performance identification, the present embodiment is more towards language
Activity recognition method also includes：According to text message and text feature information, review text information.With reference to the original of text message
Sentence and text feature information, the original statement of text message is modified or rewritten, to be conducive to language performance to recognize, drop
The data processing complex degree of low language performance identification process.
Meanwhile, to state in realization in processing procedure, the present embodiment is more mainly to use algorithm towards language performance recognition methods
It is as follows：Decision tree and rule-based algorithm (decision tree＆rule-based algorithm), random forests algorithm (random
Forest), Gradient Propulsion machine (gradient boosting machine), algorithm of support vector machine (support vector
Machine), deep learning algorithm-convolutional neural networks (convolutional neural network), deep learning are calculated
Method-Recognition with Recurrent Neural Network (recurrent neural network), the carrying out of above-mentioned algorithm is arranged in pairs or groups and adjusted, to reach more
Identification precision high.
Second aspect, the embodiment of the present invention provides a kind of many towards language performance identifying system, and with reference to Fig. 2 or Fig. 3, this is
System includes information acquisition subsystem 1, information extraction subsystem 2 and language performance recognition subsystem 3, and information acquisition subsystem 1 is used
In the current information, user portrait information and the external information that obtain user input, current information is many letters towards language performance
Breath；Information extraction subsystem 2 is used to carry out current information information extraction, obtains characteristic information；Language performance recognition subsystem 3
For according to characteristic information, user portrait information and external information, carrying out language performance identification, the language performance of user is obtained.
As shown from the above technical solution, the present embodiment is more towards language performance identifying system, can be many according to user input
Towards the current information of language performance, feature information extraction is carried out, it is to avoid enter line statement fractionation, to be conducive to according to current information
Engage in the dialogue reply.Meanwhile, the system combination user portrait information and external information carry out language performance identification, can reduce
Language performance recognizes error situation, reduces language performance False Rate.
Therefore, the present embodiment is more towards language performance identifying system, can precisely, efficiently identify the language row of user
For, user language behavior False Rate is reduced, it is favorably improved communication efficiency.
In order to further improve many accuracys and treatment effeciency towards language performance identifying system of the present embodiment, with reference to figure
3, information extraction subsystem 2 specifically for：Semantics information extraction is carried out to text message, text feature information is obtained, to sound
Information carries out sound characteristic extraction, obtains sound characteristic information, and image feature extraction is carried out to image information, obtains image feature
Information, current information includes text message, acoustic information and image information, and characteristic information includes that text feature information, sound are special
Reference ceases and image character information；Language performance recognition subsystem 3 specifically for：Believed according to text feature information, sound characteristic
Breath, image character information, user portrait information and external information, carry out language performance identification, and record instruction Activity recognition
Number, if language performance is recognized successfully, generates the language performance of the user, if language performance recognition failures, language is carried out again
Speech Activity recognition, until language performance is recognized successfully, or language performance identification number of times reaches preset value.
Here, information extraction subsystem 2 can process many information, information extraction subsystem 2 includes text feature
Module, sound characteristic module and image feature module, wherein, text feature module can process text message, sound characteristic mould
Block can process acoustic information, and image feature module can process image information, and language performance recognition subsystem 3 is special according to text
At least one of reference breath, sound characteristic information, image character information three, and combine user's portrait information and outside letter
Breath, carries out language performance identification, wherein, according to text feature information, sound characteristic information, image character information, user's portrait
Information and external information, carry out language performance identification, obtain the degree of accuracy highest of language performance.Meanwhile, in language performance identification
During failure, the information extraction subsystem 2 can also carry out repeating identification, and to obtain identification language performance, this is to be conducive to raising
The stability and Consumer's Experience of system.
Meanwhile, the present embodiment is more also to determine subsystem towards language performance identifying system including preset value, for obtaining sample
This text feature information, sample audio characteristic information, sample image characteristic information, user portrait information and external information；According to
Sample text characteristic information, sample audio characteristic information, sample image characteristic information, user portrait information and external information, enter
The identification of row language performance, obtains specimen discerning language performance；The true language performance of sample and specimen discerning language according to pre-acquiring
The similarity that words and deeds are, sets preset value.Here, the preset value determines that subsystem sets iterations according to the sample of training,
I.e. preset value, with Optimal Parameters, is favorably improved accuracy and treatment effeciency that the system is processed.
Specifically, the present embodiment is more also includes text message amendment subsystem towards language performance identifying system, for root
According to text message and text feature information, review text information.The original language of text message amendment subsystem combination text message
Sentence and text feature information, the original statement of text message is modified or rewritten, and to be conducive to language performance to recognize, is reduced
The complexity of the data processing of language performance identification process.
Finally it should be noted that：Various embodiments above is merely illustrative of the technical solution of the present invention, rather than its limitations；To the greatest extent
Pipe has been described in detail with reference to foregoing embodiments to the present invention, it will be understood by those within the art that：Its according to
The technical scheme described in foregoing embodiments can so be modified, or which part or all technical characteristic are entered
Row equivalent；And these modifications or replacement, the essence of appropriate technical solution is departed from various embodiments of the present invention technology
The scope of scheme, it all should cover in the middle of the scope of claim of the invention and specification.
Claims (8)
1. a kind of many towards language performance recognition methods, it is characterised in that including：
Obtain current information, user portrait information and the external information of user input；
Information extraction is carried out to the current information, characteristic information is obtained；
According to the characteristic information, the user portrait information and the external information, language performance identification is carried out, obtain user
Language performance.
2. it is many towards language performance recognition methods according to claim 1, it is characterised in that letter is carried out to the current information
Breath is extracted, and obtains characteristic information, is specifically included：
Semantics information extraction is carried out to text message, text feature information is obtained；
Sound characteristic extraction is carried out to acoustic information, sound characteristic information is obtained；
Image feature extraction is carried out to image information, image character information is obtained,
The current information includes the text message, the acoustic information and the image information,
The characteristic information includes the text feature information, the sound characteristic information and the image character information；
According to the characteristic information, the user portrait information and the external information, language performance identification is carried out, obtain user
Language performance, specifically include：
According to the text feature information, the sound characteristic information, the image character information, the user portrait information and
The external information, carries out language performance identification, and record instruction Activity recognition number of times, if language performance is recognized successfully, gives birth to
Into the language performance of the user, if language performance recognition failures, carry out language performance identification again, until language performance is recognized
Success, or language performance identification number of times reaches preset value.
3. many towards language performance recognition methods according to claim 2, it is characterised in that obtain text feature information it
Afterwards, before carrying out language performance identification, the method also includes：According to the text message and the text feature information, amendment
The text message.
4. it is many towards language performance recognition methods according to claim 2, it is characterised in that to recognize it language performance is carried out
Before, the method also includes：
Obtain sample text characteristic information, sample audio characteristic information, sample image characteristic information, user's portrait information and outside
Information；
According to the sample text characteristic information, the sample audio characteristic information, the sample image characteristic information, the use
Family portrait information and the external information, carry out language performance identification, obtain specimen discerning language performance；
The similarity of the true language performance of sample and the specimen discerning language performance according to pre-acquiring, sets described default
Value.
5. a kind of many towards language performance identifying system, it is characterised in that including：
Information acquisition subsystem, current information, user portrait information and external information for obtaining user input；
Information extraction subsystem, for carrying out information extraction to the current information, obtains characteristic information；
Language performance recognition subsystem, for according to the characteristic information, the user portrait information and the external information, entering
The identification of row language performance, obtains the language performance of user.
6. it is many towards language performance identifying system according to claim 5, it is characterised in that described information extracts subsystem tool
Body is used for：Semantics information extraction is carried out to text message, text feature information is obtained；Sound characteristic is carried out to acoustic information to carry
Take, obtain sound characteristic information；Image feature extraction is carried out to image information, image character information, the current information is obtained
Including the text message, the acoustic information and the image information, the characteristic information include the text feature information,
The sound characteristic information and the image character information；
The language performance recognition subsystem specifically for：According to the text feature information, the sound characteristic information, described
Image character information, the user portrait information and the external information, carry out language performance identification, and record instruction behavior is known
Other number of times, if language performance is recognized successfully, generates the language performance of the user, if language performance recognition failures, enter again
Row language performance is recognized, until language performance is recognized successfully, or language performance identification number of times reaches preset value.
7. it is many towards language performance identifying system according to claim 6, it is characterised in that the system also includes text message
Amendment subsystem, for according to the text message and the text feature information, correcting the text message.
8. it is many towards language performance identifying system according to claim 6, it is characterised in that the system also includes that preset value is true
Stator system, for obtaining sample text characteristic information, sample audio characteristic information, sample image characteristic information, user's portrait
Information and external information；According to the sample text characteristic information, the sample audio characteristic information, the sample image feature
Information, the user portrait information and the external information, carry out language performance identification, obtain specimen discerning language performance；Root
According to the true language performance of the sample of pre-acquiring and the similarity of the specimen discerning language performance, the preset value is set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710012063.2A CN106682642A (en) | 2017-01-06 | 2017-01-06 | Multi-language-oriented behavior identification method and multi-language-oriented behavior identification system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710012063.2A CN106682642A (en) | 2017-01-06 | 2017-01-06 | Multi-language-oriented behavior identification method and multi-language-oriented behavior identification system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106682642A true CN106682642A (en) | 2017-05-17 |
Family
ID=58849349
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710012063.2A Pending CN106682642A (en) | 2017-01-06 | 2017-01-06 | Multi-language-oriented behavior identification method and multi-language-oriented behavior identification system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106682642A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108021934A (en) * | 2017-11-23 | 2018-05-11 | 阿里巴巴集团控股有限公司 | The method and device of more key element identifications |
| CN109036390A (en) * | 2018-08-15 | 2018-12-18 | 四川大学 | A kind of broadcast keyword recognition method based on integrated gradient elevator |
| CN113129895A (en) * | 2021-04-20 | 2021-07-16 | 上海仙剑文化传媒股份有限公司 | Voice detection processing system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104598579A (en) * | 2015-01-14 | 2015-05-06 | 北京京东尚科信息技术有限公司 | Automatic question and answer method and system |
| CN105574133A (en) * | 2015-12-15 | 2016-05-11 | 苏州贝多环保技术有限公司 | Multi-mode intelligent question answering system and method |
| CN105895103A (en) * | 2015-12-03 | 2016-08-24 | 乐视致新电子科技（天津）有限公司 | Speech recognition method and device |
| CN106055662A (en) * | 2016-06-02 | 2016-10-26 | 竹间智能科技（上海）有限公司 | Emotion-based intelligent conversation method and system |
| CN106095833A (en) * | 2016-06-01 | 2016-11-09 | 竹间智能科技（上海）有限公司 | Human computer conversation's content processing method |
-
2017
- 2017-01-06 CN CN201710012063.2A patent/CN106682642A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104598579A (en) * | 2015-01-14 | 2015-05-06 | 北京京东尚科信息技术有限公司 | Automatic question and answer method and system |
| CN105895103A (en) * | 2015-12-03 | 2016-08-24 | 乐视致新电子科技（天津）有限公司 | Speech recognition method and device |
| CN105574133A (en) * | 2015-12-15 | 2016-05-11 | 苏州贝多环保技术有限公司 | Multi-mode intelligent question answering system and method |
| CN106095833A (en) * | 2016-06-01 | 2016-11-09 | 竹间智能科技（上海）有限公司 | Human computer conversation's content processing method |
| CN106055662A (en) * | 2016-06-02 | 2016-10-26 | 竹间智能科技（上海）有限公司 | Emotion-based intelligent conversation method and system |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108021934A (en) * | 2017-11-23 | 2018-05-11 | 阿里巴巴集团控股有限公司 | The method and device of more key element identifications |
| CN108021934B (en) * | 2017-11-23 | 2022-03-04 | 创新先进技术有限公司 | Method and device for recognizing multiple elements |
| CN109036390A (en) * | 2018-08-15 | 2018-12-18 | 四川大学 | A kind of broadcast keyword recognition method based on integrated gradient elevator |
| CN109036390B (en) * | 2018-08-15 | 2022-07-08 | 四川大学 | Broadcast keyword identification method based on integrated gradient elevator |
| CN113129895A (en) * | 2021-04-20 | 2021-07-16 | 上海仙剑文化传媒股份有限公司 | Voice detection processing system |
| CN113129895B (en) * | 2021-04-20 | 2022-12-30 | 上海仙剑文化传媒股份有限公司 | Voice detection processing system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8494853B1 (en) | Methods and systems for providing speech recognition systems based on speech recordings logs | |
| CN110838289A (en) | Awakening word detection method, device, equipment and medium based on artificial intelligence | |
| CN111477216A (en) | Training method and system for pronunciation understanding model of conversation robot | |
| CN108447471A (en) | Audio recognition method and speech recognition equipment | |
| CN110910283A (en) | Method, device, equipment and storage medium for generating legal document | |
| CN107564528B (en) | Method and equipment for matching voice recognition text with command word text | |
| CN112818680B (en) | Corpus processing method and device, electronic equipment and computer readable storage medium | |
| CN106372063A (en) | Information processing method and device and terminal | |
| KR20200119410A (en) | System and Method for Recognizing Emotions from Korean Dialogues based on Global and Local Contextual Information | |
| CN109977398A (en) | A kind of speech recognition text error correction method of specific area | |
| CN108268442A (en) | A kind of sentence Intention Anticipation method and system | |
| CN112860871B (en) | Natural language understanding model training method, natural language understanding method and device | |
| CN110503943B (en) | Voice interaction method and voice interaction system | |
| CN106682642A (en) | Multi-language-oriented behavior identification method and multi-language-oriented behavior identification system | |
| CN114120985A (en) | Pacifying interaction method, system and equipment of intelligent voice terminal and storage medium | |
| CN110931002B (en) | Man-machine interaction method, device, computer equipment and storage medium | |
| CN106776557B (en) | Emotional state memory identification method and device of emotional robot | |
| CN114373443A (en) | Speech synthesis method and apparatus, computing device, storage medium, and program product | |
| CN115345177A (en) | Intention recognition model training method and dialogue method and device | |
| CN117634471A (en) | NLP quality inspection method and computer readable storage medium | |
| CN116070642A (en) | Text emotion analysis method and related device based on expression embedding | |
| CN115691503A (en) | Voice recognition method and device, electronic equipment and storage medium | |
| CN111292741B (en) | Intelligent voice interaction robot | |
| CN114974310A (en) | Emotion recognition method and device based on artificial intelligence, computer equipment and medium | |
| CN116340510B (en) | Text classification variant recall optimization method, system, medium and equipment |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170517 |