CN105825856B - The autonomous learning method of vehicle-mounted voice identification module - Google Patents
The autonomous learning method of vehicle-mounted voice identification module Download PDFInfo
- Publication number
- CN105825856B CN105825856B CN201610321781.3A CN201610321781A CN105825856B CN 105825856 B CN105825856 B CN 105825856B CN 201610321781 A CN201610321781 A CN 201610321781A CN 105825856 B CN105825856 B CN 105825856B
- Authority
- CN
- China
- Prior art keywords
- phonetic order
- voice
- vehicle
- cloud server
- library
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 16
- 230000005484 gravity Effects 0.000 claims description 31
- 238000012544 monitoring process Methods 0.000 claims description 7
- 238000013507 mapping Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/28—Constructional details of speech recognition systems
- G10L15/30—Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
-
- 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
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/063—Training
-
- 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
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/065—Adaptation
- G10L15/07—Adaptation to the speaker
-
- 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
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/063—Training
- G10L2015/0631—Creating reference templates; Clustering
-
- 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
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/063—Training
- G10L2015/0635—Training updating or merging of old and new templates; Mean values; Weighting
- G10L2015/0636—Threshold criteria for the updating
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Electrically Operated Instructional Devices (AREA)
- Navigation (AREA)
Abstract
The present invention relates to vehicle-mounted voice control technologies, and it discloses a kind of autonomous learning methods of vehicle-mounted voice identification module, improve the versatility of speech recognition.This method, comprising the following steps: a. vehicle-mounted voice identification module acquires the corresponding phonetic order of user's operation, and uploads to cloud server；B. cloud server judges the phonetic order whether is had existed in speech recognition library, if it is present entering step c, otherwise enters step d；C. information existing for the phonetic order is fed back to vehicle-mounted voice identification module, speech recognition module downloads the comparison model of the phonetic order and user's phonetic order is prompted to can be used；D. cloud server learns the instruction.The present invention is controlled suitable for vehicle-mounted voice.
Description
Technical field
The present invention relates to vehicle-mounted voice control technologies, and in particular to a kind of autonomous learning side of vehicle-mounted voice identification module
Method.
Background technique
With the extensive use of voice technology, at present in field of vehicle control, driver is assisted frequently with voice control
The non-driving function of override vehicle, not only brings the convenience of operation, moreover it is possible to which reduction driver occurs due to wanting in driving procedure
Some functions in manual manipulation central control system and the case where divert one's attention, improve drive safety.
In the prior art, what vehicle-mounted voice identification module used is all pre-configured correlation data (i.e. speech recognition
Compare basis), in use, this data provides the benchmark of comparison, but is not easy to be modified.And since driver is each
The language use habit and dialect habit, the language that single comparison data not can guarantee each individual that people is owned by oneself make
With.
Thus vehicle-mounted voice identification module in the prior art does not have versatility, and the application is a kind of vehicle-mounted it is necessary to propose
The autonomous learning method of speech recognition module, improves the versatility of speech recognition.
Summary of the invention
The technical problems to be solved by the present invention are: proposing a kind of autonomous learning method of vehicle-mounted voice identification module, mention
The versatility of high speech recognition.
The technical solution adopted by the present invention to solve the technical problems is: the autonomous learning side of vehicle-mounted voice identification module
Method, comprising the following steps:
A. the corresponding phonetic order of vehicle-mounted voice identification module acquisition user's operation, and upload to cloud server；
B. cloud server judges the phonetic order whether is had existed in speech recognition library, if it is present entering step
Rapid c, otherwise enters step d；
C. information existing for the phonetic order is fed back to vehicle-mounted voice identification module, speech recognition module is downloaded the voice and referred to
The comparison model of order simultaneously prompts user's phonetic order can be used；
D. cloud server learns the instruction.
As advanced optimizing, in step d, the cloud server learns the instruction, specifically includes:
Cloud server first determines whether that the phonetic order whether there is in the alternative library of voice, and if it exists, then increases the language
The phonetic order is then added in the alternative library of voice, and its rate of specific gravity is initialized as by the rate of specific gravity of sound instruction if it does not exist
0, when there are other users to upload the phonetic order again, increase the rate of specific gravity of the phonetic order；
Cloud server judges the rate of specific gravity of the phonetic order when a threshold is reached, by the phonetic order from the alternative library of voice
In move in speech recognition library, as usable voice command.
As advanced optimizing, in step d, further includes:
The service condition of each phonetic order in cloud server timing/real-time monitoring speech recognition library is known in voice
In other library, some phonetic order is used, and increases its rate of specific gravity, if certain time is not used, its rate of specific gravity is reduced, when certain
When the rate of specific gravity of a phonetic order is less than threshold value, then by the phonetic order from being moved in speech recognition library in the alternative library of voice.
As advanced optimizing, in step d, further includes:
The rate of specific gravity of each phonetic order in the alternative library of cloud server timing/real-time monitoring voice, when some voice refers to
It is 0 in the specific gravity certain time of order, then deletes the phonetic order from the alternative library of voice.
As advanced optimizing, in step c, the comparison model of the phonetic order includes the phonetic order and respective operations
Mapping relations.
As advanced optimizing, in step a, the vehicle-mounted voice identification module acquires the corresponding voice of user's operation and refers to
It enables, and uploads to cloud server, refer to:
User's operation and user speech instruction are simultaneously acquired and upload cloud；If user only operates no voice
Input, then user's operation is not collected and uploads；If user only has phonetic order not have respective operations, according to local (non-
Cloud) the phonetic order comparison database of storage executes the corresponding command.
The beneficial effects of the present invention are: passing through the autonomous learning of vehicle-mounted voice identification module, so that speech recognition manipulation tool
There is versatility, server is to the monitoring of introducing specific gravity and eliminative mechanism in new instruction learning process beyond the clouds: when specific gravity reaches threshold
It is formally used as instruction in typing speech recognition library when value, the instruction specific gravity in speech recognition library is reduced to threshold value or less
When, which is moved into alternative library, when the long-term specific gravity of the instruction in alternative library is 0, deletes the instruction；It so can reduce language
Sound identifies the storage pressure in library, is also convenient for managing.
Detailed description of the invention
Fig. 1 is the autonomous learning method flow diagram of vehicle-mounted voice identification module；
Fig. 2 is flow chart of the cloud server to instruction study.
Specific embodiment
The present invention is directed to propose a kind of autonomous learning method of vehicle-mounted voice identification module, improves the general of speech recognition
Property, the present invention can configure its distinctive voice using style for each driver for everyone use habit, and
This style can be adaptive with the speech habits of driver, and step is as shown in Figure 1:
A, the corresponding phonetic order of vehicle-mounted voice identification module acquisition user's operation, and upload to cloud server；
B, cloud server judges the phonetic order whether is had existed in speech recognition library, if it does, to vehicle-mounted language
Sound identification module feeds back information existing for the phonetic order, and speech recognition module downloads the comparison model of the phonetic order and prompt
User's phonetic order can be used；If it does not exist, then cloud server learns the instruction.
In specific implementation, the corresponding phonetic order of vehicle-mounted voice identification module acquisition user's operation in step A, and upload
Refer to cloud server: user's operation and user speech instruction are simultaneously acquired and upload cloud；If user only grasps
Make no voice input, then user's operation is not collected and uploads；If user only has phonetic order not have respective operations, press
The corresponding command is executed according to the phonetic order comparison database of local (non-cloud) storage.
Cloud to the study of instruction as shown in Fig. 2,
Cloud server first determines whether that the phonetic order whether there is in the alternative library of voice, and if it exists, then increases the language
The phonetic order is then added in the alternative library of voice, and its rate of specific gravity is initialized as by the rate of specific gravity of sound instruction if it does not exist
0, when there are other users to upload the phonetic order again, increase the rate of specific gravity of the phonetic order；
Cloud server judges the rate of specific gravity of the phonetic order when a threshold is reached, by the phonetic order from the alternative library of voice
In move in speech recognition library, as usable voice command.
The service condition of each phonetic order in cloud server timing/real-time monitoring speech recognition library is known in voice
In other library, some phonetic order is used, and increases its rate of specific gravity, if certain time is not used, its rate of specific gravity is reduced, when certain
When the rate of specific gravity of a phonetic order is less than threshold value, then by the phonetic order from being moved in speech recognition library in the alternative library of voice.
The rate of specific gravity of each phonetic order in the alternative library of cloud server timing/real-time monitoring voice, when some voice refers to
It is 0 in the specific gravity certain time of order, then deletes the phonetic order from the alternative library of voice.
It should be noted that the storage library that the speech recognition library in cloud is instructed as efficient voice, provides phonetic order
Downloading, and alternative library is the base library as phonetic order study, for vehicle-mounted voice identification terminal, local voice identification
The content in library might not be completely the same with cloud speech recognition library, since the phonetic order in the speech recognition library in cloud has drop
The possibility in library, local voice identification library are actually the subset in cloud history direction library；And since local voice identifies library
In include is all the local voice command downloaded, even if the order is downgraded to the alternative library of voice by cloud even deletes this
Phonetic order will not influence user and solve user speech instruction using in a wide range of interior reasonability and private ownership.
Claims (3)
1. the autonomous learning method of vehicle-mounted voice identification module, which comprises the following steps:
A. the corresponding phonetic order of vehicle-mounted voice identification module acquisition user's operation, and upload to cloud server；
B. cloud server judges the phonetic order whether is had existed in speech recognition library, if it is present c is entered step,
Otherwise d is entered step；
C. information existing for the phonetic order is fed back to vehicle-mounted voice identification module, speech recognition module downloads the phonetic order
Comparison model simultaneously prompts user's phonetic order can be used；
D. cloud server learns the instruction:
Cloud server first determines whether that the phonetic order whether there is in the alternative library of voice, and if it exists, then increases the voice and refers to
The phonetic order is then added in the alternative library of voice, and its rate of specific gravity is initialized as 0 by the rate of specific gravity of order if it does not exist, when
When there are other users to upload the phonetic order again, increase the rate of specific gravity of the phonetic order；
Cloud server judges the rate of specific gravity of the phonetic order when a threshold is reached, which is moved from the alternative library of voice
Into speech recognition library, as usable voice command；
The service condition of each phonetic order in cloud server timing/real-time monitoring speech recognition library, in speech recognition library
In, some phonetic order is used, its rate of specific gravity is increased, if certain time is not used, its rate of specific gravity is reduced, when some language
When the rate of specific gravity of sound instruction is less than threshold value, then by the phonetic order from being moved in speech recognition library in the alternative library of voice；
The rate of specific gravity of each phonetic order in the alternative library of cloud server timing/real-time monitoring voice, when some phonetic order
It is 0 in specific gravity certain time, then deletes the phonetic order from the alternative library of voice.
2. the autonomous learning method of vehicle-mounted voice identification module as described in claim 1, which is characterized in that described in step c
The comparison model of phonetic order includes the mapping relations of the phonetic order and respective operations.
3. the autonomous learning method of vehicle-mounted voice identification module as described in claim 1, which is characterized in that described in step a
Vehicle-mounted voice identification module acquires the corresponding phonetic order of user's operation, and uploads to cloud server, refers to:
User's operation and user speech instruction are simultaneously acquired and upload cloud；If it is defeated that user only operates no voice
Enter, then user's operation is not collected and uploads；If user only has phonetic order not have respective operations, according to what is locally stored
Phonetic order comparison database executes the corresponding command.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610321781.3A CN105825856B (en) | 2016-05-16 | 2016-05-16 | The autonomous learning method of vehicle-mounted voice identification module |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610321781.3A CN105825856B (en) | 2016-05-16 | 2016-05-16 | The autonomous learning method of vehicle-mounted voice identification module |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105825856A CN105825856A (en) | 2016-08-03 |
| CN105825856B true CN105825856B (en) | 2019-11-08 |
Family
ID=56529555
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610321781.3A Active CN105825856B (en) | 2016-05-16 | 2016-05-16 | The autonomous learning method of vehicle-mounted voice identification module |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105825856B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107146623B (en) * | 2017-04-07 | 2021-03-16 | 百度在线网络技术（北京）有限公司 | Speech recognition method, device and system based on artificial intelligence |
| CN107342079A (en) * | 2017-07-05 | 2017-11-10 | 谌勋 | A kind of acquisition system of the true voice based on internet |
| CN108831462A (en) * | 2018-06-26 | 2018-11-16 | 北京奇虎科技有限公司 | Vehicle-mounted voice recognition methods and device |
| CN112154640B (en) * | 2018-07-04 | 2024-04-30 | 华为技术有限公司 | Message playing method and terminal |
| CN110288990B (en) * | 2019-06-12 | 2021-07-20 | 深圳康佳电子科技有限公司 | Voice control optimization method, storage medium and intelligent terminal |
| CN111681656A (en) * | 2020-05-23 | 2020-09-18 | 达科为（深圳）医疗设备有限公司 | Embedding box marking machine instruction input method, system, storage medium and marking machine |
| CN113223535B (en) * | 2021-03-22 | 2024-04-05 | 惠州市德赛西威汽车电子股份有限公司 | Vehicle-mounted voice skill real-time recommendation and downloading system and method |
| CN114444513A (en) * | 2022-01-28 | 2022-05-06 | 重庆长安汽车股份有限公司 | Semantic mining method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101145341A (en) * | 2006-09-04 | 2008-03-19 | 美商富迪科技股份有限公司 | Method, system and apparatus for improved voice recognition |
| CN103000175A (en) * | 2012-12-03 | 2013-03-27 | 深圳市金立通信设备有限公司 | Voice recognition method and mobile terminal |
| CN103632669A (en) * | 2012-08-20 | 2014-03-12 | 上海闻通信息科技有限公司 | A method for a voice control remote controller and a voice remote controller |
| CN103685393A (en) * | 2012-09-13 | 2014-03-26 | 大陆汽车投资（上海）有限公司 | Vehicle-borne voice control terminal, voice control system and data processing system |
| CN104575494A (en) * | 2013-10-16 | 2015-04-29 | 中兴通讯股份有限公司 | Speech processing method and terminal |
| CN104778946A (en) * | 2014-01-10 | 2015-07-15 | 中国电信股份有限公司 | Voice control method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6483680B2 (en) * | 2014-06-30 | 2019-03-13 | クラリオン株式会社 | Information processing system and in-vehicle device |
-
2016
- 2016-05-16 CN CN201610321781.3A patent/CN105825856B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101145341A (en) * | 2006-09-04 | 2008-03-19 | 美商富迪科技股份有限公司 | Method, system and apparatus for improved voice recognition |
| CN103632669A (en) * | 2012-08-20 | 2014-03-12 | 上海闻通信息科技有限公司 | A method for a voice control remote controller and a voice remote controller |
| CN103685393A (en) * | 2012-09-13 | 2014-03-26 | 大陆汽车投资（上海）有限公司 | Vehicle-borne voice control terminal, voice control system and data processing system |
| CN103000175A (en) * | 2012-12-03 | 2013-03-27 | 深圳市金立通信设备有限公司 | Voice recognition method and mobile terminal |
| CN104575494A (en) * | 2013-10-16 | 2015-04-29 | 中兴通讯股份有限公司 | Speech processing method and terminal |
| CN104778946A (en) * | 2014-01-10 | 2015-07-15 | 中国电信股份有限公司 | Voice control method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105825856A (en) | 2016-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105825856B (en) | The autonomous learning method of vehicle-mounted voice identification module | |
| CN112272819B (en) | Method and system for passively waking up user interaction device | |
| CN110047487B (en) | Wake-up method and device for vehicle-mounted voice equipment, vehicle and machine-readable medium | |
| US11043211B2 (en) | Speech recognition method, electronic device, and computer storage medium | |
| US10692503B2 (en) | Voice data processing method, apparatus and storage medium | |
| CN106328148B (en) | Natural voice recognition method, device and system based on local and cloud hybrid recognition | |
| DE102013007502A1 (en) | Computer-implemented method for automatically training a dialogue system and dialog system for generating semantic annotations | |
| Larcher et al. | I-vectors in the context of phonetically-constrained short utterances for speaker verification | |
| CN105225660B (en) | The adaptive method and system of voice system | |
| CN109410927A (en) | Offline order word parses the audio recognition method combined, device and system with cloud | |
| JP2015018238A5 (en) | ||
| JP2022095768A (en) | Method, device, apparatus, and medium for dialogues for intelligent cabin | |
| CN104282307A (en) | Method, device and terminal for awakening voice control system | |
| CN107424614A (en) | A kind of sound-groove model update method | |
| CN109584875A (en) | Voice equipment control method and device, storage medium and voice equipment | |
| CN107733762B (en) | Voice control method, device and system for smart home | |
| CN108897517B (en) | Information processing method and electronic equipment | |
| CN110503943B (en) | Voice interaction method and voice interaction system | |
| US10847154B2 (en) | Information processing device, information processing method, and program | |
| JP2019061098A5 (en) | ||
| CA2897671A1 (en) | Audio command adaptive processing system and method | |
| KR20130097307A (en) | Modeling method for learning task skill and robot using thereof | |
| CN116386610A (en) | Voice recognition method and device, vehicle-mounted terminal, server and medium | |
| Casanueva et al. | Adaptive speech recognition and dialogue management for users with speech disorders. | |
| CN115440200A (en) | Control method and control system of vehicle-mounted machine system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |