CN108847237A - continuous speech recognition method and system - Google Patents
continuous speech recognition method and system Download PDFInfo
- Publication number
- CN108847237A CN108847237A CN201810847817.0A CN201810847817A CN108847237A CN 108847237 A CN108847237 A CN 108847237A CN 201810847817 A CN201810847817 A CN 201810847817A CN 108847237 A CN108847237 A CN 108847237A
- Authority
- CN
- China
- Prior art keywords
- user
- voice
- voice messaging
- module
- chat
- 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 21
- 230000001960 triggered effect Effects 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 description 6
- 230000011218 segmentation Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 3
- 239000000243 solution Substances 0.000 description 3
- 238000012937 correction Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 208000003028 Stuttering Diseases 0.000 description 1
- 206010047924 Wheezing Diseases 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001746 injection moulding Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012360 testing method Methods 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
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- 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
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L15/222—Barge in, i.e. overridable guidance for interrupting prompts
-
- 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
- G10L15/26—Speech to text systems
-
- 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
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/223—Execution procedure of a spoken command
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)
- Telephonic Communication Services (AREA)
Abstract
The present invention relates to technical field of voice recognition, more particularly to a kind of continuous speech recognition method and system, Continuous Speech Recognition System, including speech signal analysis module and speech analysis module, speech signal analysis module is used to the voice messaging that user inputs intermittently being successively arranged in voice sequence according to voice messaging acquisition time on time list, then the corresponding voice sequence in the very short part of user is subjected to editing, the voice messaging after editing is combined into continuous voice messaging and is sent to the progress speech recognition of speech analysis module and parsing.This programme is not only suitable for can be carried out the user's use continuously spoken, and the user that can not be continuously spoken also is suitble to use.
Description
Technical field
The present invention relates to technical field of voice recognition, and in particular to a kind of continuous speech recognition method and system.
Background technique
With the development of voice technology, automatic speech recognition technology has been widely used in the every field of life,
Voice is changed into text and greatly facilitates people's lives demand, session recording is such as changed into text as meeting summary and is sent to
Personnel participating in the meeting；The recording of interview is changed into text, compiles news release etc. on this basis.But turn by voice messaging
Usually occur the presence of the place of mistake in the text converted when changing text into.
In order to convert speech information into text information, the Chinese patent document of Publication No. CN107305541A is disclosed
A kind of speech recognition text segmentation method and device, this method include：End-point detection is carried out to voice data, obtains each voice segments
And the beginning frame number of each voice segments and terminate frame number；Speech recognition is carried out to each voice segments, it is corresponding to obtain each voice segments
Identify text；Extract the segmentation feature of the corresponding identification text of each voice segments；Segmentation feature and building in advance using extraction
Segmented model, identification text corresponding to the voice data carries out segmentation detection, to determine position that needs are segmented；According to
Segmentation testing result identification text corresponding to the voice data is segmented.Above scheme can be automatically realized to identification
Text is segmented, and makes to identify that the structure of an article of text is more clear.
But above scheme is that voice collecting is carried out by way of voice segments, if user terminal is language using user
The weaker child of sound organizational capacity, the stutter patient or sick and weak old man to wheeze, they have the characteristics that one it is common, be to speak
It is possible and discontinuous when the time, but it is desultory, they say and may in short pause many times, and pause all do not have every time
Regular, the dead time also has with short, may will be in short adopting if realizing the identification of voice segments according to the dead time
Collection is divided into different voice segments, at this moment, carries out being identified as text information to each voice segments, can be due to using in the voice segments
Do not finished if family and cause speech recognition at text information error rate it is higher.
Summary of the invention
The technical issues of solution of the invention, is to provide a kind of Continuous Speech Recognition System, to solve continuously to speak
The high problem of user's text information error rate that voice messaging is identified as after input voice information.
Base case provided by the invention is：Continuous Speech Recognition System, including speech signal analysis module and voice solution
Module is analysed, the voice messaging that speech signal analysis module is used to intermittently input user is believed on time list according to voice
Breath acquisition time is successively arranged in voice sequence, the corresponding voice sequence in the very short part of user is then carried out editing, by editing
Voice messaging afterwards is combined into continuous voice messaging and is sent to the progress speech recognition of speech analysis module and parsing.
The principle of the invention lies in：After the desultory input voice information of user, speech signal analysis module is by user
The voice messaging inputted intermittently is successively arranged in voice sequence according to voice messaging acquisition time on time list, then
The corresponding voice sequence in the very short part of user is subjected to editing, the voice messaging after editing is combined into continuous voice messaging and is sent out
It gives speech analysis module and carries out speech recognition and parsing.
The advantage of the invention is that：The voice messaging of acquisition is reassembled into after continuous voice messaging and carries out voice again
Parsing, avoids the voice messaging of user's a word from just stopping the acquisition of voice messaging when inputting complete not yet, that is, solves
Only is carried out to part of speech information the problem of identification causes the text information error rate after identification to increase.With prior art phase
Than this method is applicable not only to that the user of voice messaging input, equally, the user that can continuously speak can not be carried out intermittently
Also it can be used.
It further, further include chat module and good friend's title remarks module, chat module carries out good friend's addition for user
And it chats with different good friends in different chat interfaces；When user and good friend chat, the corresponding chat circle of a good friend
Face；Good friend's title remarks module carries out different title remarks to different good friends for user.
By the setting of chat module, chat convenient for user and addition good friend and with good friend；It is standby by good friend's title
Injection molding block searches out good friend and and good friend's expansion chat convenient for user.
It further, further include that voice continuously inputs jump module, voice continuously inputs jump module and exists simultaneously for user
When switching is realized from different good friends chat between multiple chat interfaces, user speech input remarks in good friend's title remarks module
After good friend's title, the chat interface of user and the good friend are opened, text of the voice messaging of at this moment user's input after parsing
Information will be shown in the chat interface, be checked for user and user friend.
Due to that may not recognize between the good friend of user's addition, in the prior art, user needs in section at the same time
It is interior to chat from different good friends, it just needs oneself to be manually switched to the chat interface between different good friends, existing skill
The mode for switching chat interface in art is also normally the mode touched manually and realizes, may also result in chat interface when touching manually
Handoff error may will will send information in the chat interface of mistake if user relatively worries, and information is caused accidentally to send,
It makes troubles to using.The switching of chat interface, chat interface are realized in this programme in such a way that voice inputs good friend's title
Easy switching and the problem for avoiding manual switching mistake reduce the problem of information is accidentally sent.After chat interface switching, use
The voice messaging of family input will chat interface after handover carry out the text information of the corresponding parsing of the voice messaging and show, that
User can quickly after a chat interface input voice information quickly and be switched to another chat interface after
It is continuous to carry out voice messaging input.In addition, completing good friend's setting for illiterate old man and child, with the help of friend layman
Afterwards, old man and child also can very easily chat with good friend.
It further, further include trigger button, it is continuously defeated that trigger button for user triggers voice when pinning trigger button
Enter jump module start-up operation.
The setting of trigger button avoids chat when user mentions the title of another good friend when chatting with a good friend
Interface jumps.
For above-mentioned Continuous Speech Recognition System, the present invention also provides a kind of continuous speech recognition method, including it is as follows
Step：
S1, voice collecting：The voice messaging for needing to carry out speech recognition to user carries out continuous collecting；
S2, voice messaging arrange：Collected voice messaging is successively arranged on time list according to acquisition time
Then the corresponding voice sequence of user's dwell portion is carried out editing, the voice messaging after editing is combined into company by voice sequence
Continuous voice messaging；
S3, speech analysis and output：Speech recognition and parsing are carried out to the continuous speech information after combination, by voice messaging
Text information after parsing carries out display output.
In step S1, continuous acquisition is carried out to the voice messaging of user's input, i.e. desultory speak of user will not
Existing on the acquisition of voice messaging influences；In step S2, after the voice messaging of acquisition is reassembled into continuous voice messaging
The speech analysis for carrying out step S3 again, avoids the voice messaging of user's a word from just stopping voice when inputting complete not yet
The acquisition of information solves and only carries out identification to part of speech information the text information error rate after identification is caused raised to be asked
Topic.Compared with prior art, this method is applicable not only to that the user of voice messaging input, equally, energy can not be carried out intermittently
Enough users continuously to speak also can be used.
Further, in step S1, when carrying out continuous acquisition to the voice messaging of user's input, if user's pause duration reaches
User sets preset duration and at this moment restarts to be acquired the voice messaging that user is inputting.
By user oneself set pause duration preset duration, just restart after finishing a word convenient for user into
Row voice collecting avoids the too long text information for leading to parsing of voice messaging acquisition duration inputted to user from exporting the waiting time
It is too long.
Detailed description of the invention
Fig. 1 is the logic diagram of Continuous Speech Recognition System in the embodiment of the present invention one；
Fig. 2 is the implementation flow chart of continuous speech recognition method in the embodiment of the present invention one；
Fig. 3 is the logic diagram of Continuous Speech Recognition System in the embodiment of the present invention three；
Fig. 4 is the implementation flow chart of continuous speech recognition method in the embodiment of the present invention three.
Specific embodiment
Embodiment one
As shown in Figure 1：A kind of Continuous Speech Recognition System, including：Including user terminal and server.Server and user
Module is communicated terminal by wireless communication, and wireless communication module can select the Bluetooth communication mould of existing DX-BT18 model
Block.User terminal can select robot, mobile phone or it is other can be for the portable electronic device of user.
User terminal includes：
Voice acquisition module, voice messaging when for speaking in real time to user are acquired, and will collect voice letter
Breath is sent to server.
Text information output module is parsed, for receiving the text information of speech analysis module transmission, and is receiving text
Text information is shown after this information.
Voice collecting interrupt module, pause duration reaches default when the input of its voice is arranged according to own situation for user
The interruption of voice collecting is carried out when duration.For example, 4 seconds expression the words of pause of speaking usually have been finished, then preset duration is 4
Second, at this moment, voice acquisition module, which just starts to resurvey voice messaging, is sent to server.
Server includes：
Database, database is for all data informations in storage server.
Speech signal analysis module for receiving the voice messaging of voice acquisition module transmission, and is pressed on time list
It is successively arranged in voice sequence according to acquisition time, the corresponding voice sequence of user's dwell portion is then subjected to editing, by editing
Voice messaging afterwards is combined into continuous voice messaging and is sent to speech analysis module.It can be understood as unanimously inputting to user
Voice messaging record, the recorded audio for part that then user does not speak carries out editing automatically, so that recorded audio
Output is coherent.
Speech analysis module, for receiving the voice messaging of speech signal analysis module transmission, then by the voice messaging
Being parsed into text information, (speech analysis can be used the existing voice of Iflytek limited liability company and know analytic technique progress voice
Identification), and the text information after parsing is sent to text information output module.
In addition, the present embodiment also discloses a kind of continuous speech recognition as shown in Fig. 2, being directed to Continuous Speech Recognition System
Method includes the following steps：
S1, voice collecting
Voice messaging when voice acquisition module in real time speaks to user is acquired, and will collect voice messaging transmission
To server.
S2, voice messaging arrange
After speech signal analysis module in server receives the voice messaging of voice acquisition module transmission, arranged in the time
It is successively arranged in voice sequence according to acquisition time on table, the corresponding voice sequence of user's dwell portion is then subjected to editing,
Voice messaging after editing is combined into continuous voice messaging and is sent to speech analysis module.
S3, speech analysis and output
Speech analysis module, for receiving the voice messaging of speech signal analysis module transmission, then by the voice messaging
Being parsed into text information, (speech analysis can be used the existing voice of Iflytek limited liability company and know analytic technique progress voice
Identification), and the text information after parsing is sent to text information output module.Text information output module receives voice solution
After analysing the voice messaging that module is sent, display output is carried out to text information.
Embodiment two
Embodiment two and embodiment one be not, in embodiment two, in order to enable the text information after parsing can
Quickly output stops user after speech signal analysis module carries out voice sequence arrangement to voice messaging on time list
While the corresponding voice sequence editing in part groups of clips unify a voice messaging side to the voice messaging after combined into
Row parsing, i.e., be first first resolved by the voice messaging that editing is combined, after parse after the voice messaging that is combined.In addition, can also root
Quick error correction is carried out according to text information of the existing voice error correction method to parsing, for example Publication No. CN103021412A is disclosed
Audio recognition method.
Embodiment three
As shown in figure 3, embodiment three and the difference of embodiment two are that in embodiment three, user terminal further includes：
Chat module is simultaneously chatted with good friend in different chat interfaces for user progress good friend addition.User and
When good friend chats, a good friend corresponds to a chat interface, and user can send voice messaging and text information in chat interface,
The good friend of user equally can be with sending information information and voice messaging.
Good friend's title remarks module carries out remarks for title of the user to each good friend, and a good friend is one corresponding
Title.
Voice continuously inputs jump module, realizes for user's switching between multiple chat interfaces simultaneously and chats from different good friends
It when, user inputs the Yong Huyu in good friend's title remarks module after good friend's title of remarks by voice acquisition module voice
The chat interface of the good friend is opened, and text information of the voice messaging of at this moment user's input after parsing will be in chat circle
Face is shown, is checked for user and user friend.When user needs and other good friends chat, it is good that voice inputs another
The title of friend, chat interface just jump to the chat interface of this good friend from the chat interface of a upper good friend, and at this moment user is defeated
Text information of the voice messaging entered after parsing will be shown in this chat interface.For example, user becomes reconciled simultaneously
Friendly A, good friend B and good friend C chat, it is assumed that A, B, C are respectively the title of these three good friends, if user is being in the chat of A
Interface, but need to reply the information of good friend B at once, at this moment, at this moment user chats from the title of voice acquisition module input good friend B
Its interface just jumps to the chat interface of good friend B from the chat interface of good friend A, user again input voice information when, voice messaging
Text information after parsing will be shown in the chat interface with good friend B, if at this moment good friend C sends information, need to reply
The information that friendly C is sent, user can then input the title of good friend C from voice acquisition module, into the chat interface with good friend C.
Operational module, including trigger button are triggered, voice is triggered when pinning trigger button for user and continuously inputs jump
The work of revolving die block, chat interface jumps when user being avoided to mention the title of another good friend when chatting with a good friend.
As shown in figure 4, embodiment three and the difference of embodiment two also reside in, embodiment three also discloses a kind of continuous speech
Recognition methods includes the following steps：
S1, addition good friend and remarks good friend's title
User carries out the addition of good friend by chat module, and it is standby then to carry out good friend's title by good friend's title remarks module
Note.
S2, user and good friend's chat
User is chatted by chat module and good friend, and when chat, user passes through voice acquisition module and carries out voice letter
Breath input, then user speaks, and to be accustomed to analysis module according to the voice messaging of the collected user and all users universal
Habits information of speaking is analyzed to obtain the habits information of speaking of the user to the habit of speaking of the user, later speech analysis mould
Root tuber parses the voice messaging that it is inputted according to the habits information of speaking of the user, and by the text information of parsing in user
Display output is carried out with the chat interface that good friend is chatting, is checked for user and good friend.
S3, chat interface automatically switch
When other good friends send information to user, user needs timely return information, and at this moment, user can pin touching
Button is sent out, the good friend's title chatted then is inputted from voice acquisition module voice, then voice is continuously inputted and jumped
Module control chat interface jumps to the chat interface of user Yu the good friend, and user unclamps again trigger button, and at this moment, user is from language
The voice messaging of sound acquisition module input will carry out display output in the chat interface after being parsed into text information.If user is also
It need to be switched to the chat interface of other good friends, can be realized by repeating step S3.
What has been described above is only an embodiment of the present invention, and the common sense such as well known specific structure and characteristic are not made herein in scheme
Excessive description, technical field that the present invention belongs to is all before one skilled in the art know the applying date or priority date
Ordinary technical knowledge can know the prior art all in the field, and have using routine experiment hand before the date
The ability of section, one skilled in the art can improve and be implemented in conjunction with self-ability under the enlightenment that the application provides
This programme, some typical known features or known method should not become one skilled in the art and implement the application
Obstacle.It should be pointed out that for those skilled in the art, without departing from the structure of the invention, can also make
Several modifications and improvements out, these also should be considered as protection scope of the present invention, these all will not influence the effect that the present invention is implemented
Fruit and patent practicability.The scope of protection required by this application should be based on the content of the claims, the tool in specification
The records such as body embodiment can be used for explaining the content of claim.
Claims (6)
1. Continuous Speech Recognition System, it is characterised in that：Including speech signal analysis module and speech analysis module, voice messaging
The voice messaging that processing module is used to intermittently input user is successive according to voice messaging acquisition time on time list
It is arranged in voice sequence, the corresponding voice sequence in the very short part of user is then subjected to editing, by the voice messaging group after editing
It synthesizes continuous voice messaging and is sent to the progress speech recognition of speech analysis module and parsing.
2. Continuous Speech Recognition System according to claim 1, it is characterised in that：It further include chat module and good friend's title
Remarks module, chat module carry out good friend's addition for user and chat with different good friends in different chat interfaces；With
When family and good friend chat, the corresponding chat interface of a good friend；Good friend's title remarks module is for user to different good friends
Carry out different title remarks.
3. Continuous Speech Recognition System according to claim 2, it is characterised in that：It further include that voice continuously inputs and jumps mould
Block, voice continuously input jump module for user while when switching is realized from different good friends chat between multiple chat interfaces,
In good friend's title remarks module after good friend's title of remarks, the chat interface of user and the good friend are opened for user speech input,
At this moment text information of the voice messaging of user's input after parsing will show in the chat interface, for user and
Family friend checks.
4. Continuous Speech Recognition System according to claim 3, it is characterised in that：It further include trigger button, trigger button
Voice is triggered when pinning trigger button for user and continuously inputs jump module start-up operation.
5. a kind of continuous speech recognition method, includes the following steps：
S1, voice collecting：The voice messaging for needing to carry out speech recognition to user carries out continuous collecting；
S2, voice messaging arrange：Collected voice messaging is successively arranged in voice according to acquisition time on time list
Then the corresponding voice sequence of user's dwell portion is carried out editing, the voice messaging after editing is combined into continuously by sequence
Voice messaging；
S3, speech analysis and output：Speech recognition and parsing are carried out to the continuous speech information after combination, voice messaging is parsed
Text information afterwards carries out display output.
6. continuous speech recognition method according to claim 5, it is characterised in that：In step S1, to the language of user's input
When message breath carries out continuous acquisition, if user's pause duration reaches user, at this moment setting preset duration restarts to user just
It is acquired in the voice messaging of input.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810847817.0A CN108847237A (en) | 2018-07-27 | 2018-07-27 | continuous speech recognition method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810847817.0A CN108847237A (en) | 2018-07-27 | 2018-07-27 | continuous speech recognition method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108847237A true CN108847237A (en) | 2018-11-20 |
Family
ID=64192206
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810847817.0A Pending CN108847237A (en) | 2018-07-27 | 2018-07-27 | continuous speech recognition method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108847237A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109787880A (en) * | 2018-12-11 | 2019-05-21 | 平安科技（深圳）有限公司 | Voice transmission method, device, computer equipment and the storage medium at quick interface |
| CN109961787A (en) * | 2019-02-20 | 2019-07-02 | 北京小米移动软件有限公司 | Determine the method and device of acquisition end time |
| CN110660393A (en) * | 2019-10-31 | 2020-01-07 | 广东美的制冷设备有限公司 | Voice interaction method, device, equipment and storage medium |
| WO2021004236A1 (en) * | 2019-07-08 | 2021-01-14 | 深圳开立生物医疗科技股份有限公司 | Voice control method and system, device and computer-readable storage medium |
| CN113113007A (en) * | 2021-03-30 | 2021-07-13 | 北京金山云网络技术有限公司 | Voice data processing method and device, electronic equipment and storage medium |
| CN116312485A (en) * | 2023-05-23 | 2023-06-23 | 广州小鹏汽车科技有限公司 | Voice recognition method and device and vehicle |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06202689A (en) * | 1992-12-28 | 1994-07-22 | Sony Corp | Method and device for speech recognition |
| JPH07230293A (en) * | 1994-02-17 | 1995-08-29 | Sony Corp | Voice recognition device |
| DE19939102C1 (en) * | 1999-08-18 | 2000-10-26 | Siemens Ag | Speech recognition method for dictating system or automatic telephone exchange |
| CN1945563A (en) * | 2005-10-04 | 2007-04-11 | 罗伯特·博世有限公司 | Natural language processing of disfluent sentences |
| KR20080023030A (en) * | 2006-09-08 | 2008-03-12 | 한국전자통신연구원 | On-line speaker recognition method and apparatus for thereof |
| US20120065968A1 (en) * | 2010-09-10 | 2012-03-15 | Siemens Aktiengesellschaft | Speech recognition method |
| CN102662704A (en) * | 2012-03-31 | 2012-09-12 | 上海量明科技发展有限公司 | Method, terminal and system for starting instant messaging interaction interface |
| CN102903361A (en) * | 2012-10-15 | 2013-01-30 | Itp创新科技有限公司 | Instant call translation system and instant call translation method |
| CN104267922A (en) * | 2014-09-16 | 2015-01-07 | 联想(北京)有限公司 | Information processing method and electronic equipment |
| CN105139849A (en) * | 2015-07-22 | 2015-12-09 | 百度在线网络技术（北京）有限公司 | Speech recognition method and apparatus |
| CN106935253A (en) * | 2017-03-10 | 2017-07-07 | 北京奇虎科技有限公司 | The method of cutting out of audio file, device and terminal device |
| CN107146602A (en) * | 2017-04-10 | 2017-09-08 | 北京猎户星空科技有限公司 | A kind of audio recognition method, device and electronic equipment |
| CN107195303A (en) * | 2017-06-16 | 2017-09-22 | 北京云知声信息技术有限公司 | Method of speech processing and device |
| CN107293300A (en) * | 2017-08-01 | 2017-10-24 | 珠海市魅族科技有限公司 | Audio recognition method and device, computer installation and readable storage medium storing program for executing |
-
2018
- 2018-07-27 CN CN201810847817.0A patent/CN108847237A/en active Pending
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06202689A (en) * | 1992-12-28 | 1994-07-22 | Sony Corp | Method and device for speech recognition |
| JPH07230293A (en) * | 1994-02-17 | 1995-08-29 | Sony Corp | Voice recognition device |
| DE19939102C1 (en) * | 1999-08-18 | 2000-10-26 | Siemens Ag | Speech recognition method for dictating system or automatic telephone exchange |
| CN1945563A (en) * | 2005-10-04 | 2007-04-11 | 罗伯特·博世有限公司 | Natural language processing of disfluent sentences |
| KR20080023030A (en) * | 2006-09-08 | 2008-03-12 | 한국전자통신연구원 | On-line speaker recognition method and apparatus for thereof |
| US20120065968A1 (en) * | 2010-09-10 | 2012-03-15 | Siemens Aktiengesellschaft | Speech recognition method |
| CN102662704A (en) * | 2012-03-31 | 2012-09-12 | 上海量明科技发展有限公司 | Method, terminal and system for starting instant messaging interaction interface |
| CN102903361A (en) * | 2012-10-15 | 2013-01-30 | Itp创新科技有限公司 | Instant call translation system and instant call translation method |
| CN104267922A (en) * | 2014-09-16 | 2015-01-07 | 联想(北京)有限公司 | Information processing method and electronic equipment |
| CN105139849A (en) * | 2015-07-22 | 2015-12-09 | 百度在线网络技术（北京）有限公司 | Speech recognition method and apparatus |
| CN106935253A (en) * | 2017-03-10 | 2017-07-07 | 北京奇虎科技有限公司 | The method of cutting out of audio file, device and terminal device |
| CN107146602A (en) * | 2017-04-10 | 2017-09-08 | 北京猎户星空科技有限公司 | A kind of audio recognition method, device and electronic equipment |
| CN107195303A (en) * | 2017-06-16 | 2017-09-22 | 北京云知声信息技术有限公司 | Method of speech processing and device |
| CN107293300A (en) * | 2017-08-01 | 2017-10-24 | 珠海市魅族科技有限公司 | Audio recognition method and device, computer installation and readable storage medium storing program for executing |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109787880A (en) * | 2018-12-11 | 2019-05-21 | 平安科技（深圳）有限公司 | Voice transmission method, device, computer equipment and the storage medium at quick interface |
| CN109961787A (en) * | 2019-02-20 | 2019-07-02 | 北京小米移动软件有限公司 | Determine the method and device of acquisition end time |
| WO2021004236A1 (en) * | 2019-07-08 | 2021-01-14 | 深圳开立生物医疗科技股份有限公司 | Voice control method and system, device and computer-readable storage medium |
| CN110660393A (en) * | 2019-10-31 | 2020-01-07 | 广东美的制冷设备有限公司 | Voice interaction method, device, equipment and storage medium |
| CN110660393B (en) * | 2019-10-31 | 2021-12-03 | 广东美的制冷设备有限公司 | Voice interaction method, device, equipment and storage medium |
| CN113113007A (en) * | 2021-03-30 | 2021-07-13 | 北京金山云网络技术有限公司 | Voice data processing method and device, electronic equipment and storage medium |
| CN116312485A (en) * | 2023-05-23 | 2023-06-23 | 广州小鹏汽车科技有限公司 | Voice recognition method and device and vehicle |
| CN116312485B (en) * | 2023-05-23 | 2023-08-25 | 广州小鹏汽车科技有限公司 | Voice recognition method and device and vehicle |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108847237A (en) | continuous speech recognition method and system | |
| CN105100360B (en) | Call householder method and device for voice communication | |
| CN103634472B (en) | User mood and the method for personality, system and mobile phone is judged according to call voice | |
| CN103458056B (en) | Speech intention judging system based on automatic classification technology for automatic outbound system | |
| CN106710593B (en) | Method, terminal and server for adding account | |
| CN110035187A (en) | A method of realizing AI and operator attendance seamless switching in the phone | |
| US20170242847A1 (en) | Apparatus and method for translating a meeting speech | |
| KR20140105673A (en) | Supporting Method And System For communication Service, and Electronic Device supporting the same | |
| CN106713111B (en) | Processing method for adding friends, terminal and server | |
| CN111883168B (en) | Voice processing method and device | |
| CN103595852A (en) | A voice auxiliary input method and a voice auxiliary input apparatus | |
| CN102662704A (en) | Method, terminal and system for starting instant messaging interaction interface | |
| CN110266900B (en) | Method and device for identifying customer intention and customer service system | |
| CN105488026A (en) | Concerned topic reminding method and apparatus | |
| CN102640084B (en) | For Communications Interface Unit and the method for multi-user and system | |
| CN111128241A (en) | Intelligent quality inspection method and system for voice call | |
| CN107545887A (en) | Phonetic order processing method and processing device | |
| CN103856602A (en) | System and method for duplicating call | |
| EP2763136B1 (en) | Method and system for obtaining relevant information from a voice communication | |
| CN110751950A (en) | Police conversation voice recognition method and system based on big data | |
| CN109065041A (en) | A kind of voice interactive system and method based on robot | |
| CN111062221A (en) | Data processing method, data processing device, electronic equipment and storage medium | |
| EP2913822B1 (en) | Speaker recognition | |
| CN109873744A (en) | A kind of language conversion equipment | |
| CN107888745A (en) | The delet method and device of failure number in a kind of address list |
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: 20181120 |