CN107644643A - A kind of voice interactive system and method - Google Patents
A kind of voice interactive system and method Download PDFInfo
- Publication number
- CN107644643A CN107644643A CN201710889487.7A CN201710889487A CN107644643A CN 107644643 A CN107644643 A CN 107644643A CN 201710889487 A CN201710889487 A CN 201710889487A CN 107644643 A CN107644643 A CN 107644643A
- Authority
- CN
- China
- Prior art keywords
- module
- model
- identification
- voice
- text message
- 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
Landscapes
- User Interface Of Digital Computer (AREA)
Abstract
The invention discloses a kind of voice interactive system and method, including：Pre-processing module, it includes sound pretreatment module and endpoint detection module；Acoustic model repository, it includes HMM model matching unit, TDNN Model Matchings unit, HMM model database and ANN model database；Language model storehouse, it includes N Gram model databases and Rule based model databases；Identification module, it includes MFCC parameter attributes extraction unit and identification control unit；Interaction center, it includes semantic understanding module, interaction process module, response message storehouse and semantic dictionary database.The present invention utilizes monitoring of the feedback module to identification information and the feedback command of client, and by identification control unit to present customers identification information, while control and change Sound Match model and language model storehouse, optimize the interactive identification accuracy of interactive system well.
Description
Technical field
The present invention relates to interactive voice field, and in particular to a kind of voice interactive system and method.
Background technology
As in artificial intelligence Stepping into daily life, people are also increasingly deeper for the understanding of interactive voice, while to people
The requirement of voice interactive system also more and more higher in work intelligence.A kind of refrigerator of existing patent CN107146622A, interactive voice system
System, voice interactive method, computer equipment, computer-readable recording medium, wherein, voice interactive system includes：Voice collecting
Module, for being sent to speech signal collection and pretreatment, and by pretreated voice signal to sound identification module；Voice
Identification module, for pretreated voice signal to be converted into text message, and text message is respectively sent to processing mould
Block and semantic understanding module；Semantic understanding module, for text message to be converted into semantic results, and by semantic results send to
Processing module；Processing module, for according to text message and semantic results, inquiring about and combining associated information bank to perform language
Interactive voice task corresponding to sound signal.
Although existing patent CN107146622A realizes language man-machine interaction, interactive experience degree is improved.But also deposit
In following drawback：Either there is the language that significantly can not identify or not understand client when client occurs giving an irrelevant answer with system interaction
When message ceases, often customer is at one's wit's end, greatly reduces the performance of man-machine interaction.
The content of the invention
It is an object of the invention to provide a kind of voice interactive system and method, to solve what is proposed in above-mentioned background technology
Problem.
To achieve the above object, the present invention provides following technical scheme：
A kind of voice interactive system, including：
Pre-processing module, it includes sound pretreatment module and endpoint detection module, the sound pretreatment module difference
It is electrically connected with the microphone and endpoint detection module；
Acoustic model repository, it includes HMM model matching unit, TDNN Model Matchings unit, HMM model database and ANN
Model database, the HMM model matching unit are electrically connected with the HMM model database, the TDNN Model Matchings list
It is first to be electrically connected with the ANN model database；
Language model storehouse, it includes N-Gram model databases and Rule-based model databases；
Identification module, it includes MFCC parameter attributes extraction unit and identification control unit, and the MFCC parameter attributes carry
Unit is taken to be electrically connected with respectively with the identification control unit and endpoint detection module；The identification control unit respectively with it is described
N-Gram model databases, Rule-based model databases, HMM model databases, ANN model database, semantic understanding mould
Block and feedback module are electrically connected with；
Interaction center, it includes semantic understanding module, interaction process module, response message storehouse and semantic dictionary database,
The interaction process module is electrical with the semantic understanding module, feedback module, response message storehouse and voice synthetic module respectively
Connection, the voice synthetic module are electrically connected with feedback module and loudspeaker respectively.
Preferably, the sound pretreatment module is included at antialiasing filter processing, A/D converter turn and framing adding window
Reason.
Preferably, the endpoint detection module is the endpoint detection module based on frequency band variance.
Preferably, the identification control unit is ARM single-chip microcomputers, and it is integrated with sound model selection circuit above, is used
In selection HMM model matching unit and TDNN Model Matching units, while language model storehouse selection circuit is integrated with, for selecting
N-Gram model databases and ANN model database.
Preferably, the response mapping of the response message storehouse memory storage situational dialogues.
Preferably, the semantic dictionary databases storage statement semantics mapping.
Preferably, memory is provided with the feedback module, the text passed over for storing the interaction process module
The voice messaging and feedback command that this information, voice synthetic module pass over.
A kind of voice interactive method, including：
Step 1：Noisy speech is passed to the voice interactive system from microphone, in the sound pretreatment module process
It is middle that process is based on frequency band side again after antialiasing filter filtering process, the conversion of A/D converter signal and framing windowing process
Voice signal is obtained after the endpoint detection module of difference；
Step 2：Voice signal in step 1 obtains characteristic vector after the MFCC parameter attributes extraction unit,
After the characteristic vector is sent to the identification control unit, the identification control unit control sound model selection circuit and language
Speech model library selection circuit selects corresponding Model Matching and language model storehouse that characteristic vector is identified to obtain original text
Information；
Step 3：Mapping and semantic understanding mould of the original text information by the semantic dictionary database in step 2
The processing of block is identified text message；
Step 4：Mapping and interaction process module of the identification text message by the response message storehouse in step 3
Processing obtains response text message, while the interaction process module is by the original text information in step 2, step 3 kind
Identification text message and response text message pass to the feedback module；
Step 5：Response text message in step 4 is after voice synthetic module processing by response text message
It is converted into after voice messaging through loudspeaker and customer communication；
Step 6：The identification text message in original text information, step 3 in the feedback module storing step two
With the response text message in step 4, when repeating identical original text information, identification text message or initial text
This information, identification text message are feedback command, then the feedback module sends feedback signal to described control unit；
Step 7：Described control unit controls the interaction process module to walk after receiving the feedback signal in step 6
Identification text message in rapid three sends the voice synthetic module to, and identifies text to present customers by the loudspeaker
Information；
Step 8：To after customer's text exhibition information, described control unit control sound model selection circuit and sound
Model selection circuit changes Sound Match model and language model storehouse.
Compared with prior art, the beneficial effects of the invention are as follows：
The present invention utilizes monitoring of the feedback module to identification information and the feedback command of client, and passes through identification control unit
To present customers identification information, while control and change Sound Match model and language model, optimize the friendship of interactive system well
Mutual recognition correct rate.
Brief description of the drawings
Fig. 1 is the voice interactive system structural representation of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
As shown in Figure 1, a kind of speech recognition system, including pre-processing module, it includes sound pretreatment module and end
Point detection module, sound pretreatment module are electrically connected with microphone and endpoint detection module respectively, and acoustic model repository, it includes
HMM model matching unit, TDNN Model Matchings unit, HMM model database and ANN model database, HMM model matching unit
It is electrically connected with HMM model database, TDNN Model Matchings unit is electrically connected with ANN model database.Language model storehouse,
It includes N-Gram model databases and Rule-based model databases, and identification module, it includes the extraction of MFCC parameter attributes
Unit and identification control unit, MFCC parameter attributes extraction unit electrically connect with identification control unit and endpoint detection module respectively
Connect；Identification control unit respectively with N-Gram model databases, Rule-based model databases, HMM model database, ANN
Model database, semantic understanding module and feedback module are electrically connected with.Interaction center, it include semantic understanding module, interaction at
Manage module, response message storehouse and semantic dictionary database, interaction process module respectively with semantic understanding module, feedback module, should
Answer information bank and voice synthetic module is electrically connected with, voice synthetic module is electrically connected with feedback module and loudspeaker respectively.Sound
Sound pretreatment module include antialiasing filter processing, A/D converter turn and framing windowing process, endpoint detection module be based on
The endpoint detection module of frequency band variance, identification control unit are ARM single-chip microcomputers, and it is integrated with sound model selection electricity above
Road, for selecting HMM model matching unit and TDNN Model Matching units, while language model storehouse selection circuit is integrated with, used
In selection N-Gram model databases and ANN model databases.The response mapping of response message storehouse memory storage situational dialogues, language
Adopted dictionary database memory storage statement semantics mapping.Memory is provided with feedback module, for storing interaction process module transmission
The voice messaging and feedback command that text message, the voice synthetic module to come over passes over.
After noisy speech sends out voice interactive system by microphone input, pre-processing module is to noisy speech carry out sound
Sound pre-processes and end-point detection, and the voice that stays in after end-point detection obtains corresponding Characteristic Vectors after the processing of MFCC parameter attributes
Amount, the matching of recognition unit control selections HMM model and the combination of N-Gram model datas carry out the matching to characteristic vector, match
To original text information, original text information arrives identification text message after semantic understanding module carries out semantic understanding, known
Other text message obtains response text message after the processing by interaction process module, response text message passes through phonetic synthesis mould
Block synthetic speech signal, is pronounced by loudspeaker, while interaction process module passes original text information and identification text message
Feedback module is given, feedback module sends feedback information to identification control unit in real time, and identification control unit is according to feedback letter
Breath is handled it.
A kind of method of interactive voice, including：
Step 1：Noisy speech is passed to voice interactive system from microphone, by anti-during sound pretreatment module
Pass through the end points based on frequency band variance again after aliasing filter filtering process, the conversion of A/D converter signal and framing windowing process
Voice signal is obtained after detection module；
Step 2：Voice signal obtains characteristic vector after MFCC parameter attribute extraction units, and characteristic vector is sent to
After identification control unit, recognition unit control sound model selection circuit and language model storehouse selection circuit select corresponding model
Characteristic vector is identified to obtain original text information for matching and language model storehouse；
Step 3：Original text information is known by the mapping of semantic dictionary database and the processing of semantic understanding module
Other text message；
Step 4：Identification text message obtains response text by the mapping in response message storehouse and the processing of interaction process module
This information, while original text information, identification text message and response text message are passed to feedback mould by interaction process module
Block；
Step 5：Response text message is converted into voice messaging by response text message after voice synthetic module is handled
By loudspeaker and customer communication；
Step 6：Feedback module storing initial text message, identification text message and response text message, when repeating
Identical original text information, identification text message or original text information, identification text message are feedback command, then feed back
Module sends feedback signal to control unit；
Step 7：Control unit controls interaction process module to identify that text message sends language to after receiving feedback signal
Sound synthesis module, and text message is identified to present customers by loudspeaker；
Step 8：To after customer's text exhibition information, described control unit control sound model selection circuit and sound
Model selection circuit changes Sound Match model and language model storehouse.
When feedback module continuously monitors that original text information is identical three times, identification control unit will control transmission anti-
Identification information is showed user, while identification control unit by feedback instruction control feedback module by the form of display screen and voice
Control sound model selection circuit and language model storehouse selection circuit to change language model storehouse and acoustic model repository, handed over so as to improve
Mutual mass.
Although an embodiment of the present invention has been shown and described, for the ordinary skill in the art, can be with
A variety of changes, modification can be carried out to these embodiments, replace without departing from the principles and spirit of the present invention by understanding
And modification, the scope of the present invention is defined by the appended.
Claims (8)
- A kind of 1. voice interactive system, it is characterised in that including:Pre-processing module, it includes sound pretreatment module and endpoint detection module, the sound pretreatment module respectively with institute State microphone and endpoint detection module is electrically connected with；Acoustic model repository, it includes HMM model matching unit, TDNN Model Matchings unit, HMM model database and ANN model Database, the HMM model matching unit and the HMM model database are electrically connected with, the TDNN Model Matchings unit with The ANN model database is electrically connected with；Language model storehouse, it includes N-Gram model databases and Rule-based model databases；Identification module, it includes MFCC parameter attributes extraction unit and identification control unit, and the MFCC parameter attributes extraction is single Member is electrically connected with the identification control unit and endpoint detection module respectively；The identification control unit respectively with the N- Gram model databases, Rule-based model databases, HMM model database, ANN model database, semantic understanding module It is electrically connected with feedback module；Interaction center, it includes semantic understanding module, interaction process module, response message storehouse and semantic dictionary database, described Interaction process module is electrically connected with the semantic understanding module, feedback module, response message storehouse and voice synthetic module respectively, The voice synthetic module is electrically connected with feedback module and loudspeaker respectively.
- 2. a kind of voice interactive system according to claim 1, it is characterised in that the sound pretreatment module includes anti- Aliasing filter processing, A/D converter turns and framing windowing process.
- 3. a kind of voice interactive system according to claim 1, it is characterised in that the endpoint detection module is based on frequency Endpoint detection module with variance.
- 4. a kind of voice interactive system according to claim 1, it is characterised in that the identification control unit is that ARM is mono- Piece machine, and it is integrated with sound model selection circuit above, for selecting HMM model matching unit and TDNN Model Matching lists Member, while language model storehouse selection circuit is integrated with, for selecting N-Gram model databases and ANN model database.
- A kind of 5. voice interactive system according to claim 1, it is characterised in that the response message storehouse memory storage scene The response mapping of dialogue.
- A kind of 6. voice interactive system according to claim 1, it is characterised in that the semantic dictionary databases storage Statement semantics map.
- 7. a kind of voice interactive system according to claim 1, it is characterised in that storage is provided with the feedback module Device, the voice letter passed over for storing text message, voice synthetic module that the interaction process module passes over Breath and feedback command.
- A kind of a kind of 8. voice interactive method described in claim 1, it is characterised in that including：Step 1：Noisy speech is passed to the voice interactive system from microphone, is passed through during the sound pretreatment module Pass through again based on frequency band variance after crossing antialiasing filter filtering process, the conversion of A/D converter signal and framing windowing process Voice signal is obtained after endpoint detection module；Step 2：Voice signal in step 1 obtains characteristic vector after the MFCC parameter attributes extraction unit, described After characteristic vector is sent to the identification control unit, the identification control unit control sound model selection circuit and language mould Type storehouse selection circuit selects corresponding Model Matching and language model storehouse that characteristic vector is identified to obtain original text information；Step 3：The mapping and semantic understanding module of original text information in step 2 by the semantic dictionary database Processing is identified text message；Step 4：Identification text message in step 3 is by the mapping in the response message storehouse and the processing of interaction process module Obtain response text message, while the interaction process module is by the identification of original text information, step 3 kind in step 2 Text message and response text message pass to the feedback module；Step 5：Response text message in step 4 changes response text message after voice synthetic module processing Through loudspeaker and customer communication after into voice messaging；Step 6：The identification text message and step in original text information, step 3 in the feedback module storing step two Response text message in rapid four, believe when repeating identical original text information, identification text message or original text Breath, identification text message are feedback command, then the feedback module sends feedback signal to described control unit；Step 7：Described control unit controls the interaction process module by step 3 after receiving the feedback signal in step 6 In identification text message send the voice synthetic module to, and text envelope is identified to present customers by the loudspeaker Breath；Step 8：To after customer's text exhibition information, described control unit control sound model selection circuit and sound model Selection circuit changes Sound Match model and language model storehouse.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710889487.7A CN107644643A (en) | 2017-09-27 | 2017-09-27 | A kind of voice interactive system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710889487.7A CN107644643A (en) | 2017-09-27 | 2017-09-27 | A kind of voice interactive system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107644643A true CN107644643A (en) | 2018-01-30 |
Family
ID=61122605
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710889487.7A Pending CN107644643A (en) | 2017-09-27 | 2017-09-27 | A kind of voice interactive system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107644643A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108418744A (en) * | 2018-02-06 | 2018-08-17 | 国网浙江省电力有限公司信息通信分公司 | A kind of electronics seat system for promoting electrical power services quality |
| CN108932946A (en) * | 2018-06-29 | 2018-12-04 | 百度在线网络技术（北京）有限公司 | The voice interactive method and device that visitor need to service |
| CN109147768A (en) * | 2018-09-13 | 2019-01-04 | 云南电网有限责任公司 | A kind of audio recognition method and system based on deep learning |
| CN109616095A (en) * | 2018-12-12 | 2019-04-12 | 安徽讯呼信息科技有限公司 | A kind of AI intelligent voice system |
| CN109741643A (en) * | 2019-01-14 | 2019-05-10 | 博拉网络股份有限公司 | The word processing method of text-oriented big data |
| CN109961777A (en) * | 2019-02-16 | 2019-07-02 | 天津大学 | A kind of voice interactive method based on intelligent robot |
| WO2020024582A1 (en) * | 2018-07-28 | 2020-02-06 | 华为技术有限公司 | Speech synthesis method and related device |
| CN110784603A (en) * | 2019-10-18 | 2020-02-11 | 深圳供电局有限公司 | Intelligent voice analysis method and system for offline quality inspection |
| CN111081252A (en) * | 2019-12-03 | 2020-04-28 | 深圳追一科技有限公司 | Voice data processing method and device, computer equipment and storage medium |
| CN111326141A (en) * | 2018-12-13 | 2020-06-23 | 南京硅基智能科技有限公司 | Method for processing and acquiring human voice data |
| CN112397067A (en) * | 2020-11-13 | 2021-02-23 | 重庆长安工业(集团)有限责任公司 | Voice control terminal of weapon equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101075435A (en) * | 2007-04-19 | 2007-11-21 | 深圳先进技术研究院 | Intelligent chatting system and its realizing method |
| CN101281745A (en) * | 2008-05-23 | 2008-10-08 | 深圳市北科瑞声科技有限公司 | Interactive system for vehicle-mounted voice |
| CN101944359A (en) * | 2010-07-23 | 2011-01-12 | 杭州网豆数字技术有限公司 | Voice recognition method facing specific crowd |
| CN104882141A (en) * | 2015-03-03 | 2015-09-02 | 盐城工学院 | Serial port voice control projection system based on time delay neural network and hidden Markov model |
| CN106095798A (en) * | 2016-05-31 | 2016-11-09 | 北京红马传媒文化发展有限公司 | The recognition methods of sex characteristics, system and the terminal of a kind of feature based data base |
| 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 |
| WO2017156640A1 (en) * | 2016-03-18 | 2017-09-21 | Fluent.Ai Inc. | Method and device for automatically learning relevance of words in a speech recognition system |
-
2017
- 2017-09-27 CN CN201710889487.7A patent/CN107644643A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101075435A (en) * | 2007-04-19 | 2007-11-21 | 深圳先进技术研究院 | Intelligent chatting system and its realizing method |
| CN101281745A (en) * | 2008-05-23 | 2008-10-08 | 深圳市北科瑞声科技有限公司 | Interactive system for vehicle-mounted voice |
| CN101944359A (en) * | 2010-07-23 | 2011-01-12 | 杭州网豆数字技术有限公司 | Voice recognition method facing specific crowd |
| CN104882141A (en) * | 2015-03-03 | 2015-09-02 | 盐城工学院 | Serial port voice control projection system based on time delay neural network and hidden Markov model |
| WO2017156640A1 (en) * | 2016-03-18 | 2017-09-21 | Fluent.Ai Inc. | Method and device for automatically learning relevance of words in a speech recognition system |
| CN106095798A (en) * | 2016-05-31 | 2016-11-09 | 北京红马传媒文化发展有限公司 | The recognition methods of sex characteristics, system and the terminal of a kind of feature based data base |
| 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 (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108418744A (en) * | 2018-02-06 | 2018-08-17 | 国网浙江省电力有限公司信息通信分公司 | A kind of electronics seat system for promoting electrical power services quality |
| CN108932946A (en) * | 2018-06-29 | 2018-12-04 | 百度在线网络技术（北京）有限公司 | The voice interactive method and device that visitor need to service |
| CN108932946B (en) * | 2018-06-29 | 2020-03-13 | 百度在线网络技术（北京）有限公司 | Voice interaction method and device for customer-on-demand service |
| WO2020024582A1 (en) * | 2018-07-28 | 2020-02-06 | 华为技术有限公司 | Speech synthesis method and related device |
| CN109147768A (en) * | 2018-09-13 | 2019-01-04 | 云南电网有限责任公司 | A kind of audio recognition method and system based on deep learning |
| CN109616095A (en) * | 2018-12-12 | 2019-04-12 | 安徽讯呼信息科技有限公司 | A kind of AI intelligent voice system |
| CN111326141A (en) * | 2018-12-13 | 2020-06-23 | 南京硅基智能科技有限公司 | Method for processing and acquiring human voice data |
| CN109741643A (en) * | 2019-01-14 | 2019-05-10 | 博拉网络股份有限公司 | The word processing method of text-oriented big data |
| CN109961777A (en) * | 2019-02-16 | 2019-07-02 | 天津大学 | A kind of voice interactive method based on intelligent robot |
| CN110784603A (en) * | 2019-10-18 | 2020-02-11 | 深圳供电局有限公司 | Intelligent voice analysis method and system for offline quality inspection |
| CN111081252A (en) * | 2019-12-03 | 2020-04-28 | 深圳追一科技有限公司 | Voice data processing method and device, computer equipment and storage medium |
| CN112397067A (en) * | 2020-11-13 | 2021-02-23 | 重庆长安工业(集团)有限责任公司 | Voice control terminal of weapon equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107644643A (en) | A kind of voice interactive system and method | |
| CN108962255B (en) | Emotion recognition method, emotion recognition device, server and storage medium for voice conversation | |
| CN102298694A (en) | Man-machine interaction identification system applied to remote information service | |
| JP6025785B2 (en) | Automatic speech recognition proxy system for natural language understanding | |
| CN105206271A (en) | Intelligent equipment voice wake-up method and system for realizing method | |
| WO2023222088A1 (en) | Voice recognition and classification method and apparatus | |
| CN107919126A (en) | A kind of intelligent speech interactive system | |
| CN110299152A (en) | Interactive output control method, device, electronic equipment and storage medium | |
| CN104766608A (en) | Voice control method and voice control device | |
| CN111429915A (en) | Scheduling system and scheduling method based on voice recognition | |
| CN109933654A (en) | A kind of dialogue management method based on State Tree | |
| CN114944156A (en) | Article classification method, device and equipment based on deep learning and storage medium | |
| CN101867742A (en) | Television system based on sound control | |
| CN106205622A (en) | Information processing method and electronic equipment | |
| CN109994106A (en) | A kind of method of speech processing and equipment | |
| CN106653020A (en) | Multi-business control method and system for smart sound and video equipment based on deep learning | |
| CN116417003A (en) | Voice interaction system, method, electronic device and storage medium | |
| CN111128175B (en) | Spoken language dialogue management method and system | |
| CN113593565B (en) | Intelligent home device management and control method and system | |
| CN104679733B (en) | A kind of voice dialogue interpretation method, apparatus and system | |
| WO2023222090A1 (en) | Information pushing method and apparatus based on deep learning | |
| WO2016027909A1 (en) | Data structure, interactive voice response device, and electronic device | |
| CN109243458A (en) | A kind of speech recognition system for intelligent robot | |
| CN112837672B (en) | Method and device for determining conversation attribution, electronic equipment and storage medium | |
| CN112420079A (en) | Voice endpoint detection method and device, storage medium and electronic 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 | ||
| CB02 | Change of applicant information |
Address after: Room 803, room F1, two, innovation industrial park, No. 2800, new avenue of innovation, Hefei high tech Zone, Anhui Applicant after: Anhui Shuo Wei Intelligent Technology Co., Ltd. Address before: 230088, H2, building 374, two innovation industrial park, 2800 innovation Avenue, Hefei hi tech Zone, Anhui Applicant before: Anhui Shuo Wei Intelligent Technology Co., Ltd. |
|
| CB02 | Change of applicant information | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180130 |
|
| RJ01 | Rejection of invention patent application after publication |