CN106098054A - The defecator of speaker noise and method in a kind of speech recognition - Google Patents
The defecator of speaker noise and method in a kind of speech recognition Download PDFInfo
- Publication number
- CN106098054A CN106098054A CN201610410339.8A CN201610410339A CN106098054A CN 106098054 A CN106098054 A CN 106098054A CN 201610410339 A CN201610410339 A CN 201610410339A CN 106098054 A CN106098054 A CN 106098054A
- Authority
- CN
- China
- Prior art keywords
- audio signal
- speaker
- speech recognition
- signal
- user
- 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 38
- 230000005236 sound signal Effects 0.000 claims abstract description 163
- 238000012545 processing Methods 0.000 claims abstract description 33
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 230000015572 biosynthetic process Effects 0.000 abstract description 11
- 238000003786 synthesis reaction Methods 0.000 abstract description 11
- 238000012805 post-processing Methods 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 description 6
- 241000209140 Triticum Species 0.000 description 2
- 235000021307 Triticum Nutrition 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 241000238558 Eucarida Species 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K11/00—Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/16—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/175—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound
- G10K11/178—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R3/00—Circuits for transducers, loudspeakers or microphones
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Telephone Function (AREA)
Abstract
The invention discloses defecator and the method for speaker noise in a kind of speech recognition.Wherein, in described speech recognition, the defecator of speaker noise includes dio Output Modules, speaker, processing module and mike, and the first audio signal user chosen by dio Output Modules is exported to speaker;Rear speaker play described first audio signal;Post-processing module according to described first audio signal generate reverse audio signal;Mike receives described first audio signal, reverse audio signal and the voice signal of user's input afterwards, and export voice signal to be identified after carrying out synthesis process to speech database, by generating the reverse audio signal contrary with the first audio signal waveform of speaker output, eliminate the background noise of mobile terminal output itself, the definition of the voice signal of user's input when substantially increasing speech recognition, and then improve the accuracy of speech recognition.
Description
Technical field
The present invention relates to technical field of voice recognition, particularly to the defecator of speaker noise in a kind of speech recognition
And method.
Background technology
Speech recognition technology the most progressively becomes the key technology of man-machine interface, speech recognition technology and language in information technology
Sound synthetic technology combines and enables people to get rid of keyboard, is operated by voice command.The rise of mobile Internet just becomes
The most important applied environment of speech recognition, such as the Siri of Apple, domestic news fly software etc., it is possible to identify user efficiently
Voice.
On intelligent terminal, similar software nearly all can be installed at present, it is possible to user speech is converted into word, and by language
Sound mates with background data base, generates text importing, is the most directly controlled.In order to efficient identification voice, need
Want to avoid environmental noise during user input voice as far as possible.But having a kind of situation, i.e. terminal when playing music, user faces toward wheat
Gram wind is spoken, and can bring the musical sound of speaker into, cause recognition efficiency to decline to a great extent.
Thus prior art could be improved and improve.
Summary of the invention
In place of above-mentioned the deficiencies in the prior art, it is an object of the invention to provide speaker noise in speech recognition
Defecator and method, can disappear by generating the reverse audio signal contrary with the first audio signal waveform of speaker output
Except the background noise of mobile terminal output itself, the definition of the voice signal of user's input when substantially increasing speech recognition,
And then improve the accuracy of speech recognition.
In order to achieve the above object, this invention takes techniques below scheme:
The defecator of speaker noise in a kind of speech recognition, comprising:
Dio Output Modules, exports to speaker for the first audio signal user chosen;
Described speaker, is used for playing described first audio signal;
Processing module, for generating reverse audio signal according to described first audio signal;
Mike, for receiving described first audio signal, reverse audio signal and the voice signal of user's input, and closes
One-tenth exports voice signal to be identified to speech database after processing.
In described speech recognition in the defecator of speaker noise, described dio Output Modules includes:
Audio decoder, for receiving the first audio signal that user chooses, and is decoded place to described first audio signal
Reason;
D/A converter, for the first audio signal after decoded process carries out digital-to-analogue conversion, and exports to described speaker.
In described speech recognition in the defecator of speaker noise, described processing module includes:
Record unit, is used for recording described first audio signal decoding coding before treatment information;
CPU, for generating reverse audio signal according to described coding information.
In described speech recognition in the defecator of speaker noise, described coding information is every in the first audio signal
The coding that the frequency of one frame, amplitude are corresponding with tone color.
In described speech recognition in the defecator of speaker noise, described CPU is specifically for according to described first sound
Frequently in signal, the frequency of each frame is corresponding with amplitude coding generate equal in magnitude, in opposite direction with the first audio signal amplitude and
The reverse audio signal that frequency is equal.
The filter method of speaker noise in a kind of speech recognition, it comprises the steps:
A, the first audio signal user chosen by dio Output Modules export to speaker;
B, by described speaker play described first audio signal;
C, by processing module according to described first audio signal generate reverse audio signal;
D, received described first audio signal, reverse audio signal and the voice signal of user's input by mike, and close
One-tenth exports voice signal to be identified to speech database after processing.
In described speech recognition in the filter method of speaker noise, described step A includes:
A1, the first audio signal chosen by audio decoder reception user, and described first audio signal is decoded place
Reason;
A2, by D/A converter the first audio signal after decoded process carried out digital-to-analogue conversion, and export to described speaker.
In described speech recognition in the filter method of speaker noise, described step C includes:
C1, by the first audio signal decoding coding before treatment information described in recording unit records, described coding information is first
The frequency of each frame in audio signal, the coding that amplitude is corresponding with tone color;
C2, by CPU according to described coding information generate reverse audio signal.
In described speech recognition in the filter method of speaker noise, described step C2 specifically includes: by CPU according to institute
State the coding that in the first audio signal, the frequency of each frame is corresponding with amplitude and generate, side equal in magnitude with the first audio signal amplitude
To contrary and that frequency is equal reverse audio signal.
In described speech recognition in the filter method of speaker noise, also include after described step D:
E, by matching module, described voice signal to be identified is mated with speech database, and export the word of correspondence to end
End display interface.
Compared to prior art, in the speech recognition that the present invention provides in the defecator of speaker noise and method, institute
State the defecator of speaker noise in speech recognition and include dio Output Modules, speaker, processing module and mike, pass through
The first audio signal that user is chosen by dio Output Modules exports to speaker;Rear speaker play described first audio frequency letter
Number;Post-processing module according to described first audio signal generate reverse audio signal;Mike receives described first sound afterwards
Frequently signal, reverse audio signal and the voice signal of user's input, and export voice signal to be identified extremely after carrying out synthesis process
Speech database, by generating the reverse audio signal contrary with the first audio signal waveform of speaker output, eliminates mobile
The background noise of terminal output itself, the definition of the voice signal of user's input, Jin Erti when substantially increasing speech recognition
The accuracy of high speech recognition.
Accompanying drawing explanation
The structured flowchart of the defecator of speaker noise in the speech recognition that Fig. 1 provides for the present invention.
The flow chart of the filter method of speaker noise in the speech recognition that Fig. 2 provides for the present invention.
Detailed description of the invention
In view of in prior art, user plays music while noise height, recognition accuracy decline when carrying out speech recognition
Etc. shortcoming, it is an object of the invention to provide defecator and the method for speaker noise in a kind of speech recognition, can be by raw
Become the reverse audio signal contrary with the first audio signal waveform of speaker output, eliminate the background of mobile terminal output itself
Noise, the definition of the voice signal of user's input when substantially increasing speech recognition, and then improve the accuracy of speech recognition.
For making the purpose of the present invention, technical scheme and effect clearer, clear and definite, develop simultaneously embodiment pair referring to the drawings
The present invention further describes.Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not used to
Limit the present invention.
Referring to Fig. 1, in the speech recognition that the present invention provides, the defecator of speaker 20 noise includes that audio frequency exports mould
Block 10, speaker 20, processing module 30, mike 40 and matching module 50, described dio Output Modules 10 connects speaker 20
With processing module 30, described speaker 20 is also connected with described processing module 30, and described processing module 30 is also by mike 40 even
Connecing matching module 50, wherein, the first audio signal that user is chosen by described dio Output Modules 10 exports to speaker 20;Institute
State speaker 20 and play described first audio signal;Described processing module 30 generates reverse audio frequency according to described first audio signal
Signal;Described mike 40 receives described first audio signal, reverse audio signal and the voice signal of user's input, and carries out
Synthesis exports voice signal to be identified to speech database after processing;Described matching module 50 is for believing described voice to be identified
Number mate with speech database, and export the word of correspondence to terminal demonstration interface.
When being embodied as, as user open a terminal on music player software play music, now dio Output Modules 10
The first audio signal user chosen exports to speaker 20, speaker 20 play described first audio signal, i.e. play
The music file that user chooses, now processing module 30 is contrary with its waveform according to the first audio signal generation one play
Reverse audio signal in order to offset described first audio signal, afterwards when user needs to carry out speech recognition, mike 40
Receive voice signal and first audio signal of speaker 20 output of user's input, simultaneously also processing module 30 by described reversely
Audio signal also exports to mike 40, exports voice signal to be identified to voice after three kinds of audio signals are carried out synthesis process
Data base, is mated described voice signal to be identified with speech database by matching module 50, and exports the word of correspondence
To terminal demonstration interface.Owing to waveform and first audio signal of reverse audio signal are contrary, therefore can offset from speaker
First audio signal of 20 outputs, i.e. the background noise of terminal plays music so that voice signal to be identified only remains user's input
Voice signal, substantially increase the definition of user's input speech signal when listening music to see a film and improve speech recognition
Accuracy, simultaneously because user does not worries the background music impact on speech recognition accuracy, listening music or seeing a film
Shi Wuxu specially turns down background sound, and mike 40 can directly carry out phonetic entry, saves the time, to the user side of bringing
Just.
Specifically, described dio Output Modules 10 includes audio decoder 101 and D/A converter 102, described audio decoder
Device 101 connects D/A converter 102 and processing module 30, and described D/A converter 102 is also connected with speaker 20, wherein said audio frequency solution
Code device 101 receives the first audio signal that user chooses, and is decoded described first audio signal processing;Described DA changes
Device 102 carries out digital-to-analogue conversion to the first audio signal after decoded process, and exports to speaker 20.I.e. user choose
First being decoded processing to it by audio decoder 101 after one audio signal, output can be played the digital signal of software identification,
By D/A converter 102, described digital signal is converted to analogue signal the most again, then played out by speaker 20, it is achieved first
The broadcasting output of audio signal.
Further, described processing module 30 includes recording unit 301 and CPU302, and described record unit 301 connects sound
Frequently decoder 101, speaker 20 and CPU302, described CPU302 is also connected with mike 40, and wherein, described record unit 301 is remembered
Record described first audio signal decoding coding before treatment information;Described CPU302 generates reverse audio frequency according to described coding information
Signal, the processor that wherein said CPU302 can use model to be PM6650, certainly may be used without the process with identical function
Device, this is not construed as limiting by the present invention.
When described first audio signal is decoded processing by audio decoder 101, described record unit 301 is recorded
The coding information of this signal decoding processor, specifically, described coding information be frequency f1 of each frame in the first audio signal,
The coding that amplitude A 1 is corresponding with tone color B1, is generated and the first audio signal waveform phase according to described coding information by CPU302 afterwards
Anti-reverse audio signal, in order to offset described first audio signal in follow-up synthesis in processing, reduces in speech recognition
Speaker 20 noise, improves the accuracy of speech recognition.
Wherein, described CPU302 is specifically for corresponding with amplitude according to the frequency of each frame in described first audio signal
Coding generates equal in magnitude with the first audio signal amplitude, in opposite direction and that frequency is equal reverse audio signal, i.e. label
Unit 301 have recorded the coding that frequency f1 of each frame in the first audio signal, amplitude A 1 are corresponding with tone color B1, and CPU302 is according to this
The coding reverse audio signal of Infomiation Production one, the amplitude A 2 of described reverse audio signal and amplitude A 1 size of the first audio signal
Equal, reversely contrary, and frequency f1 of frequency f2 of described reverse audio signal and the first audio signal is equal, i.e. A2=A1,
F2=f1, so that the waveform of reverse audio signal and the first audio signal contrast, can offset in follow-up synthesis processes
Described first audio signal, thus reduce the background noise of speaker 20 output.
The present invention also provides for the filter method of speaker noise in a kind of speech recognition accordingly, as in figure 2 it is shown, described voice
In identification, the filter method of speaker noise comprises the steps
S100, the first audio signal user chosen by dio Output Modules export to speaker;
S200, play described first audio signal by speaker;
S300, by processing module according to described first audio signal generate reverse audio signal;
S400, received described first audio signal, reverse audio signal and the voice signal of user's input by mike, and carry out
Synthesis exports voice signal to be identified to speech database after processing.
When the present invention plays music by the music player software on opening a terminal user, dio Output Modules is by user
The the first audio signal output chosen, to speaker, is play described first audio signal by speaker, is i.e. play what user chose
Music file, now processing module generates a reverse audio frequency letter contrary with its waveform according to the first audio signal play
Number in order to offset described first audio signal, afterwards when user needs to carry out speech recognition, mike receives user's input
Voice signal and the first audio signal of speaker output, described reverse audio signal is also exported to wheat by also processing module simultaneously
Gram wind, exports voice signal to be identified to speech database after three kinds of audio signals are carried out synthesis process, will by matching module
Described voice signal to be identified mates with speech database, and exports the word of correspondence to terminal demonstration interface.Due to instead
Contrary to waveform and first audio signal of audio signal, therefore can offset the first audio signal from speaker output, i.e.
The background noise of terminal plays music so that voice signal to be identified only remains the voice signal of user's input, substantially increases use
The definition of family input speech signal when listening music to see a film and improve the accuracy of speech recognition, simultaneously because user is not
With worrying the background music impact on speech recognition accuracy, when listening music or seeing a film without specially turning down background sound,
Directly mike can be carried out phonetic entry, save the time, bring conveniently to user.
Specifically, described step S100 includes:
S101, the first audio signal chosen by audio decoder reception user, and described first audio signal is decoded
Process;
S102, by D/A converter, the first audio signal after decoded process is carried out digital-to-analogue conversion, and export to speaker.
First being decoded processing to it by audio decoder after the first audio signal that i.e. user chooses, output can be played
The digital signal of software identification, is converted to analogue signal by D/A converter by described digital signal the most again, then is entered by speaker
Row is play, it is achieved the broadcasting output of the first audio signal.
Further, described step S300 includes:
S301, by the first audio signal decoding coding before treatment information described in recording unit records, described coding information is
The frequency of each frame in one audio signal, the coding that amplitude is corresponding with tone color;
S302, by CPU according to described coding information generate reverse audio signal.
When described first audio signal is decoded processing by audio decoder, this signal under described recording unit records
The coding information of decoding processor, specifically, described coding information is frequency f1 of each frame in the first audio signal, amplitude A 1
The coding corresponding with tone color B1, is generated contrary with the first audio signal waveform reverse by CPU afterwards according to described coding information
Audio signal, in order to offset described first audio signal in follow-up synthesis in processing, the speaker reduced in speech recognition is made an uproar
Sound, improves the accuracy of speech recognition.
Wherein, described step S302 specifically includes and according to the frequency of each frame in described first audio signal and is shaken by CPU
The coding that width is corresponding generates equal in magnitude with the first audio signal amplitude, in opposite direction and that frequency is equal reverse audio signal.
The coding that i.e. frequency f1 of each frame in recording unit records the first audio signal, amplitude A 1 are corresponding with tone color B1, CPU according to
This reverse audio signal of coding Infomiation Production one, the amplitude A 2 of described reverse audio signal is big with the amplitude A 1 of the first audio signal
Little equal, reversely contrary, and frequency f1 of frequency f2 of described reverse audio signal and the first audio signal is equal, i.e. A2=
A1, f2=f1, so that the waveform of reverse audio signal and the first audio signal contrast, energy in follow-up synthesis processes
Offset described first audio signal, thus reduce the background noise of speaker output.
In sum, in the speech recognition that the present invention provides in the defecator of speaker noise and method, described voice
In identification, the defecator of speaker noise includes dio Output Modules, speaker, processing module and mike, defeated by audio frequency
Go out the first audio signal that user chooses by module to export to speaker;Rear speaker play described first audio signal;It
Post-processing module generates reverse audio signal according to described first audio signal;Mike receives described first audio frequency letter afterwards
Number, reverse audio signal and the voice signal of user's input, and export voice signal to be identified to voice after carrying out synthesis process
Data base, by generating the reverse audio signal contrary with the first audio signal waveform of speaker output, eliminates mobile terminal
The background noise of output itself, the definition of the voice signal of user's input when substantially increasing speech recognition, and then improve language
The accuracy of sound identification.
It is understood that for those of ordinary skills, can be according to technical scheme and send out
Bright design in addition equivalent or change, and all these change or replace the guarantor that all should belong to appended claims of the invention
Protect scope.
Claims (10)
1. the defecator of speaker noise in a speech recognition, it is characterised in that including:
Dio Output Modules, exports to speaker for the first audio signal user chosen;
Described speaker, is used for playing described first audio signal;
Processing module, for generating reverse audio signal according to described first audio signal;
Mike, for receiving described first audio signal, reverse audio signal and the voice signal of user's input, and closes
One-tenth exports voice signal to be identified to speech database after processing.
The defecator of speaker noise in speech recognition the most according to claim 1, it is characterised in that described audio frequency is defeated
Go out module to include:
Audio decoder, for receiving the first audio signal that user chooses, and is decoded place to described first audio signal
Reason;
D/A converter, for the first audio signal after decoded process carries out digital-to-analogue conversion, and exports to described speaker.
The defecator of speaker noise in speech recognition the most according to claim 2, it is characterised in that described process mould
Block includes:
Record unit, is used for recording described first audio signal decoding coding before treatment information;
CPU, for generating reverse audio signal according to described coding information.
The defecator of speaker noise in speech recognition the most according to claim 3, it is characterised in that described coding is believed
The coding that breath is the frequency of each frame in the first audio signal, amplitude is corresponding with tone color.
The defecator of speaker noise in speech recognition the most according to claim 4, it is characterised in that described CPU has
Body generates for the coding corresponding with amplitude according to the frequency of each frame in described first audio signal and shakes with the first audio signal
Equal in magnitude, in opposite direction and that frequency is equal reverse audio signal.
6. the filter method of speaker noise in a speech recognition, it is characterised in that comprise the steps:
A, the first audio signal user chosen by dio Output Modules export to speaker;
B, by described speaker play described first audio signal;
C, by processing module according to described first audio signal generate reverse audio signal;
D, received described first audio signal, reverse audio signal and the voice signal of user's input by mike, and close
One-tenth exports voice signal to be identified to speech database after processing.
The filter method of speaker noise in speech recognition the most according to claim 6, it is characterised in that described step A
Including:
A1, the first audio signal chosen by audio decoder reception user, and described first audio signal is decoded place
Reason;
A2, by D/A converter the first audio signal after decoded process carried out digital-to-analogue conversion, and export to described speaker.
The filter method of speaker noise in speech recognition the most according to claim 7, it is characterised in that described step C
Including:
C1, by the first audio signal decoding coding before treatment information described in recording unit records, described coding information is first
The frequency of each frame in audio signal, the coding that amplitude is corresponding with tone color;
C2, by CPU according to described coding information generate reverse audio signal.
The filter method of speaker noise in speech recognition the most according to claim 8, it is characterised in that described step C2
Specifically include: generated and the first sound according to the coding that the frequency of each frame in described first audio signal is corresponding with amplitude by CPU
Frequently the reverse audio signal that signal amplitude is equal in magnitude, in opposite direction and frequency is equal.
The filter method of speaker noise in speech recognition the most according to claim 6, it is characterised in that described step D
The most also include:
E, by matching module, described voice signal to be identified is mated with speech database, and export the word of correspondence to end
End display interface.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610410339.8A CN106098054A (en) | 2016-06-13 | 2016-06-13 | The defecator of speaker noise and method in a kind of speech recognition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610410339.8A CN106098054A (en) | 2016-06-13 | 2016-06-13 | The defecator of speaker noise and method in a kind of speech recognition |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106098054A true CN106098054A (en) | 2016-11-09 |
Family
ID=57228751
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610410339.8A Pending CN106098054A (en) | 2016-06-13 | 2016-06-13 | The defecator of speaker noise and method in a kind of speech recognition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106098054A (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106569774A (en) * | 2016-11-11 | 2017-04-19 | 青岛海信移动通信技术股份有限公司 | Method of eliminating noise, and terminal |
CN106782591A (en) * | 2016-12-26 | 2017-05-31 | 惠州Tcl移动通信有限公司 | A kind of devices and methods therefor that phonetic recognization rate is improved under background noise |
CN108091333A (en) * | 2017-12-28 | 2018-05-29 | 广东欧珀移动通信有限公司 | Sound control method and Related product |
CN108766468A (en) * | 2018-06-12 | 2018-11-06 | 歌尔科技有限公司 | A kind of intelligent sound detection method, wireless headset, TWS earphones and terminal |
CN109215664A (en) * | 2018-11-07 | 2019-01-15 | 珠海格力电器股份有限公司 | voice processing method and device |
CN110970027A (en) * | 2019-12-25 | 2020-04-07 | 上海博泰悦臻电子设备制造有限公司 | Voice recognition method, device, computer storage medium and system |
CN111316351A (en) * | 2017-11-09 | 2020-06-19 | Ask工业有限公司 | Device for generating an acoustically compensated signal |
CN111464541A (en) * | 2020-04-01 | 2020-07-28 | 重庆德信建信息技术有限公司 | Electronic government affair multifunctional intelligent hardware terminal based on biological recognition and working method |
CN111951802A (en) * | 2020-08-10 | 2020-11-17 | 山东金东数字创意股份有限公司 | Visual expression system and method based on AI voice recognition |
WO2021008534A1 (en) * | 2019-07-15 | 2021-01-21 | 华为技术有限公司 | Voice wakeup method and electronic device |
CN112270930A (en) * | 2020-10-22 | 2021-01-26 | 江苏峰鑫网络科技有限公司 | Method for voice recognition conversion |
CN113077779A (en) * | 2021-03-10 | 2021-07-06 | 泰凌微电子(上海)股份有限公司 | Noise reduction method and device, electronic equipment and storage medium |
CN115497474A (en) * | 2022-09-13 | 2022-12-20 | 广东浩博特科技股份有限公司 | Control method based on voice recognition |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100311405A1 (en) * | 2009-06-05 | 2010-12-09 | Hong Fu Jin Precision Industry (Shenzhen) Co., Ltd | Mobile phone and method for controlling volume of mobile phone |
CN102625006A (en) * | 2011-01-31 | 2012-08-01 | 深圳三石科技有限公司 | Method and system for synchronization and alignment of echo cancellation data and audio communication equipment |
CN203014945U (en) * | 2012-12-06 | 2013-06-19 | 联美新视信息科技(上海)有限公司 | Mobile terminal |
CN104505099A (en) * | 2014-12-08 | 2015-04-08 | 北京云知声信息技术有限公司 | Method and equipment for removing known interference in voice signal |
CN204836533U (en) * | 2015-08-18 | 2015-12-02 | 聚熵信息技术(上海)有限公司 | Acoustic control intelligence audio amplifier with echo elimination function |
CN204928959U (en) * | 2015-09-16 | 2015-12-30 | 天翼爱音乐文化科技有限公司 | Mobile terminal's music broadcast system |
CN105551479A (en) * | 2015-12-22 | 2016-05-04 | 努比亚技术有限公司 | Mobile terminal noise reduction method and apparatus thereof |
-
2016
- 2016-06-13 CN CN201610410339.8A patent/CN106098054A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100311405A1 (en) * | 2009-06-05 | 2010-12-09 | Hong Fu Jin Precision Industry (Shenzhen) Co., Ltd | Mobile phone and method for controlling volume of mobile phone |
CN102625006A (en) * | 2011-01-31 | 2012-08-01 | 深圳三石科技有限公司 | Method and system for synchronization and alignment of echo cancellation data and audio communication equipment |
CN203014945U (en) * | 2012-12-06 | 2013-06-19 | 联美新视信息科技(上海)有限公司 | Mobile terminal |
CN104505099A (en) * | 2014-12-08 | 2015-04-08 | 北京云知声信息技术有限公司 | Method and equipment for removing known interference in voice signal |
CN204836533U (en) * | 2015-08-18 | 2015-12-02 | 聚熵信息技术(上海)有限公司 | Acoustic control intelligence audio amplifier with echo elimination function |
CN204928959U (en) * | 2015-09-16 | 2015-12-30 | 天翼爱音乐文化科技有限公司 | Mobile terminal's music broadcast system |
CN105551479A (en) * | 2015-12-22 | 2016-05-04 | 努比亚技术有限公司 | Mobile terminal noise reduction method and apparatus thereof |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106569774B (en) * | 2016-11-11 | 2020-07-10 | 青岛海信移动通信技术股份有限公司 | Method and terminal for removing noise |
CN106569774A (en) * | 2016-11-11 | 2017-04-19 | 青岛海信移动通信技术股份有限公司 | Method of eliminating noise, and terminal |
CN106782591A (en) * | 2016-12-26 | 2017-05-31 | 惠州Tcl移动通信有限公司 | A kind of devices and methods therefor that phonetic recognization rate is improved under background noise |
CN111316351B (en) * | 2017-11-09 | 2024-05-28 | Ask工业有限公司 | Device for generating an acoustic compensation signal |
CN111316351A (en) * | 2017-11-09 | 2020-06-19 | Ask工业有限公司 | Device for generating an acoustically compensated signal |
CN108091333A (en) * | 2017-12-28 | 2018-05-29 | 广东欧珀移动通信有限公司 | Sound control method and Related product |
CN108766468A (en) * | 2018-06-12 | 2018-11-06 | 歌尔科技有限公司 | A kind of intelligent sound detection method, wireless headset, TWS earphones and terminal |
CN109215664B (en) * | 2018-11-07 | 2021-02-05 | 珠海格力电器股份有限公司 | Voice processing method and device |
CN109215664A (en) * | 2018-11-07 | 2019-01-15 | 珠海格力电器股份有限公司 | voice processing method and device |
WO2021008534A1 (en) * | 2019-07-15 | 2021-01-21 | 华为技术有限公司 | Voice wakeup method and electronic device |
CN110970027A (en) * | 2019-12-25 | 2020-04-07 | 上海博泰悦臻电子设备制造有限公司 | Voice recognition method, device, computer storage medium and system |
CN110970027B (en) * | 2019-12-25 | 2023-07-25 | 博泰车联网科技(上海)股份有限公司 | Voice recognition method, device, computer storage medium and system |
CN111464541A (en) * | 2020-04-01 | 2020-07-28 | 重庆德信建信息技术有限公司 | Electronic government affair multifunctional intelligent hardware terminal based on biological recognition and working method |
CN111951802A (en) * | 2020-08-10 | 2020-11-17 | 山东金东数字创意股份有限公司 | Visual expression system and method based on AI voice recognition |
CN112270930A (en) * | 2020-10-22 | 2021-01-26 | 江苏峰鑫网络科技有限公司 | Method for voice recognition conversion |
CN113077779A (en) * | 2021-03-10 | 2021-07-06 | 泰凌微电子(上海)股份有限公司 | Noise reduction method and device, electronic equipment and storage medium |
CN115497474A (en) * | 2022-09-13 | 2022-12-20 | 广东浩博特科技股份有限公司 | Control method based on voice recognition |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106098054A (en) | The defecator of speaker noise and method in a kind of speech recognition | |
WO2018121275A1 (en) | Method and apparatus for error connection of voice recognition in smart hardware device | |
JP7244665B2 (en) | end-to-end audio conversion | |
CN106782591B (en) | Device and method for improving speech recognition rate under background noise | |
O’Shaughnessy | Automatic speech recognition: History, methods and challenges | |
CN105448294A (en) | Intelligent voice recognition system for vehicle equipment | |
CN109448725A (en) | A kind of interactive voice equipment awakening method, device, equipment and storage medium | |
CN204836533U (en) | Acoustic control intelligence audio amplifier with echo elimination function | |
US20100198583A1 (en) | Indicating method for speech recognition system | |
CN205354646U (en) | Intelligence speech recognition system for mobile unit | |
CN104079699A (en) | Bluetooth earphone and control method thereof | |
CN106297790A (en) | The voiceprint service system of robot and service control method thereof | |
CN106098078A (en) | A kind of audio recognition method that may filter that speaker noise and system thereof | |
CN207603881U (en) | A kind of intelligent sound wireless sound box | |
CN110853606A (en) | Sound effect configuration method and device and computer readable storage medium | |
Liao et al. | Formosa speech recognition challenge 2020 and Taiwanese across Taiwan corpus | |
US20190152061A1 (en) | Motion control method and device, and robot with enhanced motion control | |
CN105446123A (en) | Voice intelligent alarm clock | |
Pollak et al. | Accuracy of MP3 speech recognition under real-word conditions: Experimental study | |
CN104992710A (en) | Method for finding article based on intelligent device and intelligent device | |
CN204926573U (en) | Intelligent robot of auxiliary exercise mandarin | |
CN202026434U (en) | Voice conversion STB (set top box) | |
KR102217292B1 (en) | Method, apparatus and computer-readable recording medium for improving a set of at least one semantic units by using phonetic sound | |
US7139708B1 (en) | System and method for speech recognition using an enhanced phone set | |
CN108172241B (en) | Music recommendation method and music recommendation system based on intelligent terminal |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161109 |
|
RJ01 | Rejection of invention patent application after publication |