CN109817199A - A kind of audio recognition method of fan speech control system - Google Patents
A kind of audio recognition method of fan speech control system Download PDFInfo
- Publication number
- CN109817199A CN109817199A CN201910004240.1A CN201910004240A CN109817199A CN 109817199 A CN109817199 A CN 109817199A CN 201910004240 A CN201910004240 A CN 201910004240A CN 109817199 A CN109817199 A CN 109817199A
- Authority
- CN
- China
- Prior art keywords
- electric fan
- operating mode
- uproar
- control system
- speech
- 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
Abstract
The invention discloses a kind of audio recognition methods of fan speech control system, comprising the following steps: S1: the working condition of electric fan being divided into low wind noise operating mode according to the electric fan blade revolving speed critical value of setting and high wind is made an uproar operating mode；S2: the voice data that acquisition electric fan is collected into when high wind makes an uproar operating mode is subject to adaptive noise reduction processing and machine training study, obtains being suitable for dedicated voice model of the electric fan when high wind makes an uproar operating mode；S3: electric fan is made an uproar under operating mode in high wind, after speech control system collects wave audio data, is carried out adaptive noise reduction processing to wave audio data, is retained user speech and instruct and dedicated voice model is used to be identified；S4: the working condition for the recognition result control electric fan that electric fan speech control system is instructed according to user speech.The present invention in the higher situation of flabellum revolving speed, can overcome wind make an uproar to user speech instruction interference, improve speech recognition accuracy.
Description
Technical field
The present invention relates to a kind of sound control method technical field, the voice of especially a kind of fan speech control system is known
Other method.
Background technique
Electric fan with voice control function can be by acquiring the voice command of user come the work to electric fan
State is regulated and controled, and the property easy to use of electric fan is substantially increased；Electric fan has the function of flabellum rotational speed regulation mostly, i.e., logical
The revolving speed of motor is overregulated to realize different grades of wind speed, air quantity, but with the promotion of flabellum revolving speed, can also make electric wind
The wind of fan is made an uproar increase.
Therefore in the prior art with the electric fan of voice control function, in the higher situation of flabellum revolving speed, voice is known
The user speech instruction received during not can make an uproar with the wind of electric fan to be superimposed, so that the voice signal that electric fan receives
Signal-to-noise ratio numerical value is too low, and the speech recognition accuracy for eventually leading to electric fan is lower.
Summary of the invention
For overcome the deficiencies in the prior art, the present invention provides a kind of speech recognition sides of fan speech control system
Method, in the higher situation of flabellum revolving speed, can overcome the wind of electric fan make an uproar to user speech instruction interference, improve electric fan
Speech recognition accuracy.
The technical solution adopted by the present invention to solve the technical problems is:
A kind of audio recognition method of fan speech control system, comprising the following steps:
S1: the working condition of electric fan is divided by low wind noise operating mode according to the electric fan blade revolving speed critical value of setting
It makes an uproar operating mode with high wind；
S2: the voice data that is collected into when high wind makes an uproar operating mode of acquisition electric fan, be subject to adaptive noise reduction processing and
Machine training study, obtains being suitable for dedicated voice model of the electric fan when high wind makes an uproar operating mode；
S3: electric fan is made an uproar under operating mode in high wind, after speech control system collects wave audio data, to sound sound
Frequency retains user speech and instructs and dedicated voice model is used to be identified according to adaptive noise reduction processing is carried out；
S4: the working condition for the recognition result control electric fan that electric fan speech control system is instructed according to user speech.
As a further improvement of the above technical scheme:
In step S2, the voice data that electric fan is collected into low wind noise operating mode is also acquired, machine training is subject to
Study obtains being suitable for Default sound model of the electric fan in low wind noise operating mode；
In step S3, electric fan is under low wind noise operating mode, after speech control system collects wave audio data, directly
It connects and wave audio data is identified using Default sound model.
Compared with prior art, the beneficial effects of the present invention are:
A kind of audio recognition method of fan speech control system provided by the present invention is made an uproar shape for electric fan difference wind
State develops different speech recognition modelings, and the different wind for being applied to electric fan are made an uproar in state:
High wind is in fan to make an uproar under operating mode, and noise reduction process is carried out to voice data and is carried out with dedicated voice model
Matching can effectively inhibit wind to make an uproar the influence to speech recognition, improve the accuracy rate of speech recognition；
In the case where fan is in low wind noise operating mode, the wave audio data being collected into directly are used into Default sound model
It is identified, not only saves cpu resource, and the loss of detail in user speech instruction can be caused to avoid noise reduction process,
Therefore it is also beneficial to guarantee the accuracy rate of speech recognition.
Specific embodiment
The present invention is further described combined with specific embodiments below.
A kind of audio recognition method of fan speech control system provided by the present embodiment, comprising the following steps:
S1: the working condition of electric fan is divided by low wind noise operating mode according to the electric fan blade revolving speed critical value of setting
It makes an uproar operating mode with high wind.
S2: the voice data that is collected into when high wind makes an uproar operating mode of acquisition electric fan, be subject to adaptive noise reduction processing and
Machine training study, obtains being suitable for dedicated voice model of the electric fan when high wind makes an uproar operating mode；
The voice data that electric fan is collected into low wind noise operating mode is also acquired, is subject to machine training study, obtains
Suitable for Default sound model of the electric fan in low wind noise operating mode.
S3: electric fan is made an uproar under operating mode in high wind, after speech control system collects wave audio data, to sound sound
Frequency retains user speech and instructs and dedicated voice model is used to be identified according to adaptive noise reduction processing is carried out；
Electric fan is under low wind noise operating mode, after speech control system collects wave audio data, directlys adopt silent
Recognize speech model to identify wave audio data.
S4: the working condition for the recognition result control electric fan that electric fan speech control system is instructed according to user speech.
In conclusion a kind of audio recognition method of fan speech control system provided by the present embodiment, to electric fan
Making an uproar in high wind, the progress of the voice command voice signal under state targetedly develop by speech model, passes through and acquires the high wind of electric fan
The voice data being collected under state of making an uproar, and it is subject to adaptive noise reduction processing and machine training study, it obtains being suitable for electric fan
It makes an uproar the dedicated voice model under state in high wind；A variety of adaptive noise reduction Processing Algorithms and machine have been disclosed in the prior art
Training learning method, thus the specific method and content of study are trained without repeating adaptive noise reduction processing and machine herein.
When electric fan be in high wind make an uproar state when, start dedicated voice model, by the collected sound of speech control system
Audio data carries out that wind is inhibited to make an uproar processing by adaptive noise reduction algorithm, and retains user speech and instruct, and according to the language of identification
Sound instruction control electric fan works；This method can effectively promote the signal-to-noise ratio of phonetic order, make an uproar with electric fan in high wind
Dedicated voice model under state forms higher matching degree, is conducive to the phonetic order for correctly identifying user.
It makes an uproar when electric fan is in calm or when low wind noise state, disables dedicated voice model, using Default sound model, and
Adaptive noise reduction algorithm is closed, thus to save cpu resource；Since adaptive noise reduction algorithm is while inhibiting wind to make an uproar, often
The details ingredient in user speech instruction can be lost, weakens the accuracy rate of speech recognition, therefore in the low wind noise state of electric fan
Not applicable noise reduction process and dedicated voice model down.
Better embodiment of the invention is specifically described above, certainly, the present invention can also using with it is upper
The different form of embodiment is stated, those skilled in the art are made equivalent on the premise of without prejudice to spirit of the invention
Transformation or corresponding change, all should belong in protection scope of the present invention.
Claims (2)
1. a kind of audio recognition method of fan speech control system, which comprises the following steps:
S1: the working condition of electric fan is divided by low wind noise operating mode and height according to the electric fan blade revolving speed critical value of setting
Wind is made an uproar operating mode；
S2: the voice data that acquisition electric fan is collected into when high wind makes an uproar operating mode is subject to adaptive noise reduction processing and machine
Training study obtains being suitable for dedicated voice model of the electric fan when high wind makes an uproar operating mode；
S3: electric fan is made an uproar under operating mode in high wind, after speech control system collects wave audio data, to wave audio number
According to adaptive noise reduction processing is carried out, retains user speech and instruct and dedicated voice model is used to be identified；
S4: the working condition for the recognition result control electric fan that electric fan speech control system is instructed according to user speech.
2. a kind of audio recognition method of fan speech control system according to claim 1, it is characterised in that:
In step S2, the voice data that electric fan is collected into low wind noise operating mode is also acquired, is subject to machine training study,
It obtains being suitable for Default sound model of the electric fan in low wind noise operating mode；
In step S3, electric fan after speech control system collects wave audio data, is directly adopted under low wind noise operating mode
Wave audio data are identified with Default sound model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910004240.1A CN109817199A (en) | 2019-01-03 | 2019-01-03 | A kind of audio recognition method of fan speech control system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910004240.1A CN109817199A (en) | 2019-01-03 | 2019-01-03 | A kind of audio recognition method of fan speech control system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109817199A true CN109817199A (en) | 2019-05-28 |
Family
ID=66603970
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910004240.1A Pending CN109817199A (en) | 2019-01-03 | 2019-01-03 | A kind of audio recognition method of fan speech control system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109817199A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110265053A (en) * | 2019-06-29 | 2019-09-20 | 联想(北京)有限公司 | Signal de-noising control method, device and electronic equipment |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101354887A (en) * | 2007-07-25 | 2009-01-28 | 通用汽车公司 | Ambient noise injection for use in speech recognition |
| US20090271188A1 (en) * | 2008-04-24 | 2009-10-29 | International Business Machines Corporation | Adjusting A Speech Engine For A Mobile Computing Device Based On Background Noise |
| CN102192172A (en) * | 2010-03-08 | 2011-09-21 | 艾美特电器（深圳）有限公司 | Voice control electric fan |
| CN103208284A (en) * | 2012-01-17 | 2013-07-17 | 通用汽车环球科技运作有限责任公司 | Method and system for using sound related vehicle information to enhance speech recognition |
| CN103514878A (en) * | 2012-06-27 | 2014-01-15 | 北京百度网讯科技有限公司 | Acoustic modeling method and device, and speech recognition method and device |
| US20150025881A1 (en) * | 2013-07-19 | 2015-01-22 | Audience, Inc. | Speech signal separation and synthesis based on auditory scene analysis and speech modeling |
| CN104358694A (en) * | 2014-10-28 | 2015-02-18 | 苏州古尚电器科技有限公司 | Intelligent multifunctional electrical fan |
| CN106328152A (en) * | 2015-06-30 | 2017-01-11 | 芋头科技（杭州）有限公司 | Automatic identification and monitoring system for indoor noise pollution |
| CN107257996A (en) * | 2015-03-26 | 2017-10-17 | 英特尔公司 | The method and system of environment sensitive automatic speech recognition |
| CN107464564A (en) * | 2017-08-21 | 2017-12-12 | 腾讯科技（深圳）有限公司 | voice interactive method, device and equipment |
| CN108172220A (en) * | 2018-02-22 | 2018-06-15 | 成都启英泰伦科技有限公司 | A kind of novel voice denoising method |
| CN108983684A (en) * | 2017-05-31 | 2018-12-11 | 贺州思通信息技术有限公司 | A kind of lathe safety control system based on voice recognition |
-
2019
- 2019-01-03 CN CN201910004240.1A patent/CN109817199A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101354887A (en) * | 2007-07-25 | 2009-01-28 | 通用汽车公司 | Ambient noise injection for use in speech recognition |
| US20090271188A1 (en) * | 2008-04-24 | 2009-10-29 | International Business Machines Corporation | Adjusting A Speech Engine For A Mobile Computing Device Based On Background Noise |
| CN102192172A (en) * | 2010-03-08 | 2011-09-21 | 艾美特电器（深圳）有限公司 | Voice control electric fan |
| CN103208284A (en) * | 2012-01-17 | 2013-07-17 | 通用汽车环球科技运作有限责任公司 | Method and system for using sound related vehicle information to enhance speech recognition |
| CN103514878A (en) * | 2012-06-27 | 2014-01-15 | 北京百度网讯科技有限公司 | Acoustic modeling method and device, and speech recognition method and device |
| US20150025881A1 (en) * | 2013-07-19 | 2015-01-22 | Audience, Inc. | Speech signal separation and synthesis based on auditory scene analysis and speech modeling |
| CN104358694A (en) * | 2014-10-28 | 2015-02-18 | 苏州古尚电器科技有限公司 | Intelligent multifunctional electrical fan |
| CN107257996A (en) * | 2015-03-26 | 2017-10-17 | 英特尔公司 | The method and system of environment sensitive automatic speech recognition |
| CN106328152A (en) * | 2015-06-30 | 2017-01-11 | 芋头科技（杭州）有限公司 | Automatic identification and monitoring system for indoor noise pollution |
| CN108983684A (en) * | 2017-05-31 | 2018-12-11 | 贺州思通信息技术有限公司 | A kind of lathe safety control system based on voice recognition |
| CN107464564A (en) * | 2017-08-21 | 2017-12-12 | 腾讯科技（深圳）有限公司 | voice interactive method, device and equipment |
| CN108172220A (en) * | 2018-02-22 | 2018-06-15 | 成都启英泰伦科技有限公司 | A kind of novel voice denoising method |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110265053A (en) * | 2019-06-29 | 2019-09-20 | 联想(北京)有限公司 | Signal de-noising control method, device and electronic equipment |
| CN110265053B (en) * | 2019-06-29 | 2022-04-19 | 联想(北京)有限公司 | Signal noise reduction control method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102779526B (en) | Pitch extraction and correcting method in speech signal | |
| EP1635329A3 (en) | System, method and program for sound source classification | |
| CN106448654A (en) | Robot speech recognition system and working method thereof | |
| WO2011070972A1 (en) | Voice recognition system, voice recognition method and voice recognition program | |
| CN104517607A (en) | Speed-controlled appliance and method of filtering noise therein | |
| CN105679312A (en) | Phonetic feature processing method of voiceprint identification in noise environment | |
| CN106098076A (en) | A kind of based on dynamic noise estimation time-frequency domain adaptive voice detection method | |
| SG140445A1 (en) | Method and apparatus for automatically recognizing audio data | |
| CN109817199A (en) | A kind of audio recognition method of fan speech control system | |
| CN103594094A (en) | Self-adaptive spectral subtraction real-time speech enhancement | |
| CN107039033A (en) | A kind of speech synthetic device | |
| CN106887233A (en) | Audio data processing method and system | |
| CN107102713A (en) | It is a kind of to reduce the method and device of power consumption | |
| CN101510423A (en) | Pronunciation detection method and apparatus | |
| CN108986844B (en) | Speech endpoint detection method based on speaker speech characteristics | |
| CN106439967A (en) | Method and system for increasing voice recognition rate of kitchen ventilator | |
| CN104952446A (en) | Digital building presentation system based on voice interaction | |
| CN107274895A (en) | A kind of speech recognition apparatus and method | |
| CN110988137A (en) | Abnormal sound detection system and method based on time-frequency domain characteristics | |
| CN109841208A (en) | A kind of sound enhancement method applied in range hood | |
| CN106971715A (en) | A kind of speech recognition equipment applied to robot | |
| CN111524537B (en) | Cough and sneeze identification method aiming at real-time voice flow | |
| CN111626093B (en) | Method for identifying related bird species of power transmission line based on sound power spectral density | |
| CN106887226A (en) | Speech recognition algorithm based on artificial intelligence recognition | |
| CN109814716B (en) | Movement intention decoding method based on dynamic surface electromyographic signals |
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: 20190528 |