CN105824857A - Voice search method, device and terminal - Google Patents
Voice search method, device and terminal Download PDFInfo
- Publication number
- CN105824857A CN105824857A CN201510012659.3A CN201510012659A CN105824857A CN 105824857 A CN105824857 A CN 105824857A CN 201510012659 A CN201510012659 A CN 201510012659A CN 105824857 A CN105824857 A CN 105824857A
- Authority
- CN
- China
- Prior art keywords
- recognition result
- local recognition
- terminal
- local
- result
- 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 33
- 230000005540 biological transmission Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 210000000582 semen Anatomy 0.000 description 5
- 238000000605 extraction Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect 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
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)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a voice search method which comprises the following steps: acquiring voice recording data of a user by a terminal; performing local recognition on the voice recording data by the terminal so as to obtain a local recognition result; performing validity judgment on the local recognition result by the terminal according to preset grammars and similarity thresholds; if the terminal confirms that the local recognition result is valid, confirming the local recognition result as a voice search result by the terminal; if the terminal confirms that the local recognition result is invalid, transmitting the voice recording data to a cloud server by the terminal, receiving a cloud recognition result which is transmitted from the cloud server and is obtained through cloud recognition of the voice recording data, and confirming that the cloud recognition result is the voice search result. Meanwhile, the embodiment of the invention further discloses a device and a terminal.
Description
Technical field
The present invention relates to information search field, particularly relate to a kind of voice search method, device and terminal.
Background technology
Along with the lifting of hardware performance and the becoming stronger day by day of operation system function of intelligent terminal, the quantity of information that intelligent terminal comprises is the most increasing so that it is extremely difficult that user searches for the information oneself wanted from substantial amounts of information.At present, the way of search the most easily that intelligent terminal uses is phonetic search, i.e. the voice messaging of user's input is carried out speech recognition, and obtains, according to the result of speech recognition, the information that user wants from the bulk information that intelligent terminal stores.But, the hunting zone of voice search method of the prior art is limited by the locally stored data of intelligent terminal.
Summary of the invention
For solving above-mentioned technical problem, embodiment of the present invention expectation provides a kind of voice search method, device and terminal, expands the phonetic search scope of terminal.
The technical scheme is that and be achieved in that:
First aspect, the embodiment of the present invention provides a kind of voice search method, including: terminal obtains the recording data of user；Described terminal carries out local identification to described recording data, obtains local recognition result；Described terminal carries out Effective judgement according to the grammer pre-set and similarity threshold to described local recognition result；If described terminal determines that described local recognition result is effective, the most described terminal determines that described local recognition result is phonetic search result；If described terminal determines that described local recognition result is invalid, described recording data is sent to cloud server by the most described terminal, the high in the clouds that carries out described recording data receiving the transmission of described cloud server identifies the high in the clouds recognition result obtained, and determines that described high in the clouds recognition result is described phonetic search result.
Further, described terminal carries out Effective judgement according to the grammer pre-set and similarity threshold to described local recognition result and specifically includes: if in the range of described grammer and the similarity of described local recognition result is greater than or equal to described similarity threshold for described local recognition result, it is determined that described local recognition result is effective；If at least one existed in following three kinds of situations: described local recognition result be local recognition result empty, described not in the range of described grammer, the similarity of described local recognition result is less than described similarity threshold, it is determined that described local recognition result is invalid.
Further, determine in described terminal and also include after described local recognition result is phonetic search result: described terminal, according to the data in local data base, performs the operation corresponding with described phonetic search result.
Further, determine in described terminal and also include after described high in the clouds recognition result is described phonetic search result: described terminal asks the data corresponding with described phonetic search result to described cloud server, receive the described data that described cloud server sends, and perform the operation corresponding with described phonetic search result.
Second aspect, the embodiment of the present invention provides a kind of voice searching device, including: acquiring unit, local recognition unit, judging unit, high in the clouds recognition unit and determine unit, wherein: described acquiring unit, for obtaining the recording data of user；Described local recognition unit, the described recording data for obtaining described acquiring unit carries out local identification, obtains local recognition result；Described judging unit, carries out Effective judgement for the described local recognition result obtained described local recognition unit according to the grammer pre-set and similarity threshold；Described high in the clouds recognition unit, for when described judging unit determines that described local recognition result is invalid, being sent by described recording data to cloud server, the high in the clouds that carries out described recording data receiving the transmission of described cloud server identifies the high in the clouds recognition result obtained；Described determine unit, for when described judging unit determines that described local recognition result is effective, determining that described local recognition result is phonetic search result；When described judging unit determines that described local recognition result is invalid, determine that the described high in the clouds recognition result that described high in the clouds recognition unit obtains is described phonetic search result.
Further, described judging unit specifically for: if in the range of described grammer and the similarity of described local recognition result is greater than or equal to described similarity threshold for the described local recognition result that obtains of described local recognition unit, it is determined that described local recognition result is effective；If at least one existed in following three kinds of situations: described local recognition result be local recognition result empty, described not in the range of described grammer, the similarity of described local recognition result is less than described similarity threshold, it is determined that described local recognition result is invalid.
Further, described device also includes: the first performance element, for described determine that unit determines that described local recognition result is phonetic search result after, according to the data in local data base, perform the operation corresponding with described phonetic search result.
Further, described device also includes: the second performance element, for described determine that unit determines that described high in the clouds recognition result is described phonetic search result after, to the data that the request of described cloud server is corresponding with described phonetic search result, receive the described data that described cloud server sends, and perform the operation corresponding with described phonetic search result.
The third aspect, the embodiment of the present invention provides a kind of terminal, including the voice searching device described in second aspect.
Embodiments provide a kind of voice search method, device and terminal, the recording data of user is obtained by terminal, recording data is carried out local identification and obtains local recognition result, again local recognition result is carried out Effective judgement, if the local recognition result that terminal obtains is invalid, then recording data is sent to cloud server, the high in the clouds that carries out this recording data receiving cloud server transmission identifies the high in the clouds recognition result obtained, and determine that high in the clouds recognition result is phonetic search result, so, this locality is identified and high in the clouds identifies and combines, expand the hunting zone of the phonetic search of terminal.
Accompanying drawing explanation
The schematic flow sheet of a kind of voice search method that Fig. 1 provides for the embodiment of the present invention；
The schematic flow sheet of the another kind of voice search method that Fig. 2 provides for the embodiment of the present invention；
Fig. 3 for the embodiment of the present invention provide a kind of by phonetic search and the schematic flow sheet of the voice search method as a example by playing music；
The structural representation of a kind of voice searching device that Fig. 4 provides for the embodiment of the present invention；
The structural representation of the another kind of voice searching device that Fig. 5 provides for the embodiment of the present invention.
Detailed description of the invention
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 clearly and completely described.
Fig. 1 shows a kind of voice search method that the embodiment of the present invention provides, and with reference to Fig. 1, the method includes:
S101: terminal obtains the recording data of user；
It should be noted that terminal can obtain the recording data of user in several ways, such as, by opening the sound-recording function of terminal, the voice of typing user, obtain recording data.
S102: terminal carries out local identification to recording data, obtains local recognition result；
You need to add is that, terminal carries out local identification to recording data can use multiple voice recognition methods, for example, terminal can use pattern matching method that recording data carries out local identification, voice messaging corresponding to data during terminal prestores local data base specifically, voice messaging is carried out feature extraction, obtain characteristic information, characteristic information is stored in template base as template, when needs carry out local identification to recording data, the characteristic information of recording data is carried out similarity-rough set with each template in template base successively, using similarity soprano as local recognition result.
In embodiments of the present invention, terminal carries out local identification to recording data may obtain multiple local recognition result, thus needs each local recognition result is all carried out Effective judgement, to obtain more local recognition result.
S103: terminal carries out Effective judgement according to the grammer pre-set and similarity threshold to local recognition result；
Wherein, grammer, be terminal support according to self conventional operation, use scene and user language custom etc. define, such as, " broadcasting " of music player is operated by terminal according to self, the grammer of the forms such as " song playing Wang Fei ", " broadcasting Semen Phaseoli " is set, or is accustomed to according to the term of user, the grammer of forms such as " I want to listen Semen Phaseoli " is set.
Similarity threshold, it it is the characteristic parameter similarity threshold value with the data in grammar file of terminal recording data set in advance, wherein, grammar file is by the data genaration in the local data base of terminal, including the details of Various types of data, such as, include corresponding with the song title in music data, album name, artist--name respectively song title grammer subfile, album name grammer subfile, artist--name grammer subfile.
Exemplarily, terminal carries out Effective judgement according to the grammer pre-set and similarity threshold to local recognition result and specifically includes:
If local recognition result is in the range of grammer and the similarity of local recognition result is greater than or equal to similarity threshold, it is determined that this this locality recognition result is effective；If at least one existed in following three kinds of situations: local recognition result be recognition result empty, local not in the range of grammer, the similarity of local recognition result is less than similarity threshold, it is determined that this this locality recognition result is invalid.
For example, if in recording data, the voice of user's typing is ambiguous, then local recognition result may be empty, so that it is determined that this this locality recognition result is invalid；If local recognition result is not empty, but the voice of user's typing does not conforms to grammer, such as, the voice of user's typing is " finding song Semen Phaseoli ", and grammer does not arrange such statement, then local recognition result is not in the range of grammer, so that it is determined that this this locality recognition result is invalid；If the grammatical voice of user's typing, such as, " broadcasting Semen Phaseoli ", but terminal not storage Semen Phaseoli this song is bent, then and the similarity of local recognition result is less than the similarity threshold pre-set, so that it is determined that this this locality recognition result is invalid.
If terminal determines that local recognition result is effective, then perform step S104: terminal determines that local recognition result is phonetic search result.
If terminal determines that local recognition result is invalid, then perform step S105: recording data is sent to cloud server by terminal, the high in the clouds that carries out this recording data receiving cloud server transmission identifies the high in the clouds recognition result obtained, and determines that high in the clouds recognition result is phonetic search result.
It should be noted that local identification is that recording data is identified by terminal according to locally stored data message, its recognition speed is fast, but identification range only limits the data of terminal storage；And cloud server carries out high in the clouds identification to recording data, utilizing rich in natural resources on the Internet, such as the substantial amounts of language identification other model of model harmony knowledge, recording data can be carried out larger range of speech recognition, recognition accuracy is higher.
If it is to be appreciated that terminal identifies through this locality, it is thus achieved that effective local recognition result, illustrate that in terminal, storage has user to want the relevant information of search, then can directly determine that local recognition result is phonetic search result；If terminal identifies through this locality, obtain invalid local recognition result, illustrate that not storing user in terminal wants the relevant information of search, it is thus desirable to recording data is sent to cloud server, to obtain larger range of high in the clouds recognition result, combine it can thus be seen that the embodiment of the present invention is identified by the high in the clouds that local identification terminal carried out and cloud server are carried out, extend the hunting zone of the existing phonetic search of terminal.
Exemplarily, with reference to Fig. 2, after step s 104, also include:
S106: terminal, according to the data in local data base, performs the operation corresponding with phonetic search result.
Exemplarily, with continued reference to Fig. 2, after step S105, also include:
S107: terminal asks the data corresponding with phonetic search result to cloud server, receives the above-mentioned data that cloud server sends, and performs the operation corresponding with phonetic search result.
Such as, in step S106 and step S107, if phonetic search result is a piece of music title, then perform the operation corresponding with phonetic search result and include playing this music, delete this music, edit this musical designation etc..
Fig. 3 show that the embodiment of the present invention provides by phonetic search the voice search method as a example by playing music, with reference to Fig. 3, the method includes:
S301: terminal arranges similarity threshold and grammer；
Wherein, grammer can include that+object " song title " " play " in predicate, and+object " song title " etc. " wanted to listen " in predicate.
S302: terminal obtains the recording data of user and " plays as boundless as the sea and the sky "；
S303: terminal carries out local identification to this recording data, obtains local recognition result；
S304: terminal carries out Effective judgement according to the grammer pre-set and similarity threshold to local recognition result；
S305: terminal determines that local recognition result is invalid；
Recording data " is play as boundless as the sea and the sky " and is sent to cloud server by S306: terminal；
Recording data " is play as boundless as the sea and the sky " and is carried out high in the clouds identification by S307: cloud server, and the high in the clouds recognition result obtained is " playing as boundless as the sea and the sky ", and is sent to terminal；
S308: terminal receives this high in the clouds recognition result and " plays as boundless as the sea and the sky ", and determines that " playing as boundless as the sea and the sky " is phonetic search result；
S309: terminal, according to phonetic search result, to the music file of cloud server request " as boundless as the sea and the sky ", receives the music file of " as boundless as the sea and the sky " that cloud server sends, and plays " as boundless as the sea and the sky ".
A kind of voice search method that the embodiment of the present invention provides, the recording data of user is obtained by terminal, recording data is carried out local identification and obtains local recognition result, again local recognition result is carried out Effective judgement, if the local recognition result that terminal obtains is invalid, then recording data is sent to cloud server, the high in the clouds that carries out this recording data receiving cloud server transmission identifies the high in the clouds recognition result obtained, and determine that high in the clouds recognition result is phonetic search result, so, this locality is identified and high in the clouds identifies and combines, expand the hunting zone of the phonetic search of terminal.
Fig. 4 shows the structural representation of a kind of voice searching device that the embodiment of the present invention provides, with reference to Fig. 4, this voice searching device 40 includes: acquiring unit 401, local recognition unit 402, judging unit 403, high in the clouds recognition unit 404 and determine unit 405, wherein:
Acquiring unit 401, for obtaining the recording data of user；
Local recognition unit 402, the recording data for obtaining acquiring unit 401 carries out local identification, obtains local recognition result；
Judging unit 403, carries out Effective judgement for the local recognition result obtained local recognition unit 402 according to the grammer pre-set and similarity threshold；
High in the clouds recognition unit 404, for when judging unit 403 determines that local recognition result is invalid, sends recording data to cloud server, and the high in the clouds that carries out recording data receiving cloud server transmission identifies the high in the clouds recognition result obtained；
Determine unit 405, for when judging unit 403 determines that local recognition result is effective, determining that local recognition result is phonetic search result；When judging unit 403 determines that local recognition result is invalid, determine that the high in the clouds recognition result that high in the clouds recognition unit 404 obtains is phonetic search result.
Exemplarily, it is judged that unit 403 specifically for:
If the local recognition result that local recognition unit 402 obtains is in the range of grammer and the similarity of local recognition result is greater than or equal to similarity threshold, it is determined that local recognition result is effective；If at least one existed in following three kinds of situations: local recognition result be recognition result empty, local not in the range of grammer, the similarity of local recognition result is less than similarity threshold, it is determined that local recognition result is invalid.
Exemplarily, with reference to Fig. 5, this device also includes:
First performance element 406, for, after determining that unit 405 determines that local recognition result is phonetic search result, according to the data in local data base, performing the operation corresponding with phonetic search result.
Second performance element 407, for after determining that unit 405 determines that high in the clouds recognition result is phonetic search result, to the data that cloud server request is corresponding with phonetic search result, receive the data that cloud server sends, and perform the operation corresponding with phonetic search result.
A kind of terminal that the embodiment of the present invention additionally provides, this terminal includes the voice searching device described in any of the above-described embodiment.
It should be noted that terminal can include that smart mobile phone, panel computer etc. can carry out the smart machine of speech recognition, this is not particularly limited by the embodiment of the present invention.
Those skilled in the art are it should be appreciated that embodiments of the invention can be provided as method, system or computer program.Therefore, the form of the embodiment in terms of the present invention can use hardware embodiment, software implementation or combine software and hardware.And, the present invention can use the form at one or more upper computer programs implemented of computer-usable storage medium (including but not limited to disk memory and optical memory etc.) wherein including computer usable program code.
The present invention is to describe with reference to method, equipment (system) and the flow chart of computer program according to embodiments of the present invention and/or block diagram.It should be understood that can be by the flow process in each flow process in computer program instructions flowchart and/or block diagram and/or square frame and flow chart and/or block diagram and/or the combination of square frame.These computer program instructions can be provided to produce a machine to the processor of general purpose computer, special-purpose computer, Embedded Processor or other programmable data processing device so that the instruction performed by the processor of computer or other programmable data processing device is produced for realizing the device of function specified in one flow process of flow chart or multiple flow process and/or one square frame of block diagram or multiple square frame.
These computer program instructions may be alternatively stored in and can guide in the computer-readable memory that computer or other programmable data processing device work in a specific way, the instruction making to be stored in this computer-readable memory produces the manufacture including command device, and this command device realizes the function specified in one flow process of flow chart or multiple flow process and/or one square frame of block diagram or multiple square frame.
These computer program instructions also can be loaded in computer or other programmable data processing device, make to perform sequence of operations step on computer or other programmable devices to produce computer implemented process, thus the instruction performed on computer or other programmable devices provides the step of the function specified in one flow process of flow chart or multiple flow process and/or one square frame of block diagram or multiple square frame for realization.
The above, only presently preferred embodiments of the present invention, it is not intended to limit protection scope of the present invention.
Claims (9)
1. a voice search method, it is characterised in that including:
Terminal obtains the recording data of user；
Described terminal carries out local identification to described recording data, obtains local recognition result；
Described terminal carries out Effective judgement according to the grammer pre-set and similarity threshold to described local recognition result；
If described terminal determines that described local recognition result is effective, the most described terminal determines that described local recognition result is phonetic search result；If described terminal determines that described local recognition result is invalid, described recording data is sent to cloud server by the most described terminal, the high in the clouds that carries out described recording data receiving the transmission of described cloud server identifies the high in the clouds recognition result obtained, and determines that described high in the clouds recognition result is described phonetic search result.
Method the most according to claim 1, it is characterised in that described terminal carries out Effective judgement according to the grammer pre-set and similarity threshold to described local recognition result and specifically includes:
If described local recognition result is in the range of described grammer and the similarity of described local recognition result is greater than or equal to described similarity threshold, it is determined that described local recognition result is effective；If at least one existed in following three kinds of situations: described local recognition result be local recognition result empty, described not in the range of described grammer, the similarity of described local recognition result is less than described similarity threshold, it is determined that described local recognition result is invalid.
Method the most according to claim 1, it is characterised in that determine in described terminal and also include after described local recognition result is phonetic search result:
Described terminal, according to the data in local data base, performs the operation corresponding with described phonetic search result.
Method the most according to claim 1, it is characterised in that determine in described terminal and also include after described high in the clouds recognition result is described phonetic search result:
Described terminal asks the data corresponding with described phonetic search result to described cloud server, receives the described data that described cloud server sends, and performs the operation corresponding with described phonetic search result.
5. a voice searching device, it is characterised in that including: acquiring unit, local recognition unit, judging unit, high in the clouds recognition unit and determine unit, wherein:
Described acquiring unit, for obtaining the recording data of user；
Described local recognition unit, the described recording data for obtaining described acquiring unit carries out local identification, obtains local recognition result；
Described judging unit, carries out Effective judgement for the described local recognition result obtained described local recognition unit according to the grammer pre-set and similarity threshold；
Described high in the clouds recognition unit, for when described judging unit determines that described local recognition result is invalid, being sent by described recording data to cloud server, the high in the clouds that carries out described recording data receiving the transmission of described cloud server identifies the high in the clouds recognition result obtained；
Described determine unit, for when described judging unit determines that described local recognition result is effective, determining that described local recognition result is phonetic search result；When described judging unit determines that described local recognition result is invalid, determine that the described high in the clouds recognition result that described high in the clouds recognition unit obtains is described phonetic search result.
Device the most according to claim 5, it is characterised in that described judging unit specifically for:
If the described local recognition result that described local recognition unit obtains is in the range of described grammer and the similarity of described local recognition result is greater than or equal to described similarity threshold, it is determined that described local recognition result is effective；If at least one existed in following three kinds of situations: described local recognition result be local recognition result empty, described not in the range of described grammer, the similarity of described local recognition result is less than described similarity threshold, it is determined that described local recognition result is invalid.
Device the most according to claim 5, it is characterized in that, also include: the first performance element, for described determine that unit determines that described local recognition result is phonetic search result after, according to the data in local data base, perform the operation corresponding with described phonetic search result.
Device the most according to claim 5, it is characterized in that, also include: the second performance element, for described determine that unit determines that described high in the clouds recognition result is described phonetic search result after, to the data that the request of described cloud server is corresponding with described phonetic search result, receive the described data that described cloud server sends, and perform the operation corresponding with described phonetic search result.
9. a terminal, it is characterised in that include the voice searching device described in claim 5 to 8.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510012659.3A CN105824857A (en) | 2015-01-08 | 2015-01-08 | Voice search method, device and terminal |
| PCT/CN2015/094350 WO2016110156A1 (en) | 2015-01-08 | 2015-11-11 | Voice search method and apparatus, terminal and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510012659.3A CN105824857A (en) | 2015-01-08 | 2015-01-08 | Voice search method, device and terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105824857A true CN105824857A (en) | 2016-08-03 |
Family
ID=56355488
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510012659.3A Pending CN105824857A (en) | 2015-01-08 | 2015-01-08 | Voice search method, device and terminal |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN105824857A (en) |
| WO (1) | WO2016110156A1 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106653013A (en) * | 2016-09-30 | 2017-05-10 | 北京奇虎科技有限公司 | Speech recognition method and device |
| CN112562679A (en) * | 2020-11-26 | 2021-03-26 | 浪潮金融信息技术有限公司 | Offline voice interaction method, device and medium |
| CN115394300A (en) * | 2022-10-28 | 2022-11-25 | 广州小鹏汽车科技有限公司 | Voice interaction method, voice interaction device, vehicle and readable storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708865A (en) * | 2012-04-25 | 2012-10-03 | 北京车音网科技有限公司 | Method, device and system for voice recognition |
| CN103219005A (en) * | 2013-04-28 | 2013-07-24 | 北京云知声信息技术有限公司 | Speech recognition method and device |
| CN103247291A (en) * | 2013-05-07 | 2013-08-14 | 华为终端有限公司 | Updating method, device, and system of voice recognition device |
| CN103440867A (en) * | 2013-08-02 | 2013-12-11 | 安徽科大讯飞信息科技股份有限公司 | Method and system for recognizing voice |
| CN103871408A (en) * | 2012-12-14 | 2014-06-18 | 联想(北京)有限公司 | Method and device for voice identification and electronic equipment |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102543071B (en) * | 2011-12-16 | 2013-12-11 | 安徽科大讯飞信息科技股份有限公司 | Voice recognition system and method used for mobile equipment |
| CN102970427A (en) * | 2012-11-16 | 2013-03-13 | 广东欧珀移动通信有限公司 | Method for playing songs by mobile phone |
| CN103956168A (en) * | 2014-03-29 | 2014-07-30 | 深圳创维数字技术股份有限公司 | Voice recognition method and device, and terminal |
-
2015
- 2015-01-08 CN CN201510012659.3A patent/CN105824857A/en active Pending
- 2015-11-11 WO PCT/CN2015/094350 patent/WO2016110156A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708865A (en) * | 2012-04-25 | 2012-10-03 | 北京车音网科技有限公司 | Method, device and system for voice recognition |
| CN103871408A (en) * | 2012-12-14 | 2014-06-18 | 联想(北京)有限公司 | Method and device for voice identification and electronic equipment |
| CN103219005A (en) * | 2013-04-28 | 2013-07-24 | 北京云知声信息技术有限公司 | Speech recognition method and device |
| CN103247291A (en) * | 2013-05-07 | 2013-08-14 | 华为终端有限公司 | Updating method, device, and system of voice recognition device |
| CN103440867A (en) * | 2013-08-02 | 2013-12-11 | 安徽科大讯飞信息科技股份有限公司 | Method and system for recognizing voice |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106653013A (en) * | 2016-09-30 | 2017-05-10 | 北京奇虎科技有限公司 | Speech recognition method and device |
| CN106653013B (en) * | 2016-09-30 | 2019-12-20 | 北京奇虎科技有限公司 | Voice recognition method and device |
| CN112562679A (en) * | 2020-11-26 | 2021-03-26 | 浪潮金融信息技术有限公司 | Offline voice interaction method, device and medium |
| CN115394300A (en) * | 2022-10-28 | 2022-11-25 | 广州小鹏汽车科技有限公司 | Voice interaction method, voice interaction device, vehicle and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2016110156A1 (en) | 2016-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9805733B2 (en) | Method and apparatus for connecting service between user devices using voice | |
| EP3195310B1 (en) | Keyword detection using speaker-independent keyword models for user-designated keywords | |
| EP3251115B1 (en) | Updating language understanding classifier models for a digital personal assistant based on crowd-sourcing | |
| US9875741B2 (en) | Selective speech recognition for chat and digital personal assistant systems | |
| US20170263253A1 (en) | System and method of performing automatic speech recognition using local private data | |
| KR20160006464A (en) | Method and apparatus for recognizing multi-level speech | |
| CN102024013A (en) | System and method for activating functions including a first function and a second function | |
| CN110244941B (en) | Task development method and device, electronic equipment and computer readable storage medium | |
| CN103699530A (en) | Method and equipment for inputting texts in target application according to voice input information | |
| KR20190046631A (en) | System and method for natural language processing | |
| CN112165647B (en) | Audio data processing method, device, equipment and storage medium | |
| CN110956955B (en) | Voice interaction method and device | |
| JPWO2011121649A1 (en) | Voice recognition device | |
| US20200218760A1 (en) | Music search method and device, server and computer-readable storage medium | |
| KR20200011198A (en) | Method, apparatus and computer program for providing interaction message | |
| KR20200050104A (en) | Method for providing Text-To-Speech service and relay server for the same | |
| CN104053048A (en) | Method and device for video localization | |
| CN105824857A (en) | Voice search method, device and terminal | |
| CN110942765A (en) | Method, device, server and storage medium for constructing corpus | |
| RU2014122484A (en) | SYSTEM AND METHOD FOR PROCESSING USER SPEECH REQUEST | |
| JP6481643B2 (en) | Audio processing system and audio processing method | |
| CN110931014A (en) | Speech recognition method and device based on regular matching rule | |
| CN105895098A (en) | Play control method and device | |
| CN107967308B (en) | Intelligent interaction processing method, device, equipment and computer storage medium | |
| CN103928024A (en) | Voice query method and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160803 |
|
| WD01 | Invention patent application deemed withdrawn after publication |