CN106409297A - Voice recognition method - Google Patents
Voice recognition method Download PDFInfo
- Publication number
- CN106409297A CN106409297A CN201610906925.1A CN201610906925A CN106409297A CN 106409297 A CN106409297 A CN 106409297A CN 201610906925 A CN201610906925 A CN 201610906925A CN 106409297 A CN106409297 A CN 106409297A
- Authority
- CN
- China
- Prior art keywords
- digital information
- voice
- eigenvalue
- identification
- recognition method
- 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 15
- 238000000605 extraction Methods 0.000 claims abstract description 4
- 238000002203 pretreatment Methods 0.000 claims 1
- 238000005070 sampling Methods 0.000 claims 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification 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
- G10L15/26—Speech to text systems
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
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)
- Computer Vision & Pattern Recognition (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a voice recognition method comprising the following steps that a recognized analog voice signal is digitalized; digital information is preprocessed; feature selection is performed on the preprocessed digital information, the feature values of voice required to be recognized are selected, and information including the feature values is extracted from the preprocessed digital information; the feature values required to be recognized are mapped to a feature space from an object space in the feature value extraction process; the digital information of the feature values required to be recognized is reserved in the preprocessed digital information, and other digital information is eliminated so that the digital information required to be recognized is obtained; and the digital information required to be recognized is restored into the analog voice signal. According to the computer-based voice recognition method, the voice is digitalized, feature selection is performed on the digitalized voice, and the voice is recognized through the feature values so that the recognition accuracy and the recognition speed can be greatly enhanced.
Description
Technical field
The present invention relates to voice technology field, particularly a kind of digitized audio recognition method.
Background technology
Voice, as analogue signal, starts appearance in recent years and carries out speech recognition by computer, but is difficult to logical at present
Cross computer and be digitized identifying, computer digital process is all high than traditional mode in processing speed, accuracy,
There is presently no the means by digitized processing.
Content of the invention
For solving above-mentioned technical problem, the invention provides a kind of audio recognition method, it comprises the following steps：
Identified analog voice signal is digitized, becomes and turn to the digital information being suitable to computer disposal；
Described digital information is carried out with pretreatment, removes the interference information being mixed into and reduce deformation or distortion；
Pretreated digital information is carried out with feature selection, selects to need the eigenvalue of the voice of identification, in pretreatment
The information comprising described eigenvalue is extracted in digital information afterwards；
The eigenvalue needing identification is mapped to feature space so that each needs from object space by characteristics extraction process
One of eigenvalue available feature space of identification point represents；
By the described digital information by retaining the eigenvalue needing identification in pretreated digital information, other are counted
Word information rejects the digital information obtaining needs identification；
The described digital information needing identification is carried out being reduced into analog voice signal, the analog voice letter after described reduction
Number it is the final voice needing identification.
It is preferred that the digitized process of described voice is voice signal to be digitized sample.
It is preferred that described eigenvalue includes the amplitude of sound, frequency.
The invention has the advantages that：
The computer based audio recognition method that the present invention provides passes through voice digitization, by digitized language
Sound carries out feature selection, by eigenvalue, voice is identified, and recognition accuracy is all greatly improved with recognition speed.
Certainly, the arbitrary product implementing the present invention it is not absolutely required to reach all the above advantage simultaneously.
Specific embodiment
Below in conjunction with the embodiment of the present invention, the technical scheme in the present invention is clearly and completely described it is clear that institute
The embodiment of description is only a part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention,
All other embodiment that those of ordinary skill in the art are obtained under the premise of not making creative work, broadly falls into this
The scope of bright protection.
Embodiments provide a kind of audio recognition method, it comprises the following steps：
Identified analog voice signal is digitized, becomes and turn to the digital information being suitable to computer disposal；
Described digital information is carried out with pretreatment, removes the interference information being mixed into and reduce deformation or distortion；
Pretreated digital information is carried out with feature selection, selects to need the eigenvalue of the voice of identification, in pretreatment
The information comprising described eigenvalue is extracted in digital information afterwards；
The eigenvalue needing identification is mapped to feature space so that each needs from object space by characteristics extraction process
One of eigenvalue available feature space of identification point represents；
By the described digital information by retaining the eigenvalue needing identification in pretreated digital information, other are counted
Word information rejects the digital information obtaining needs identification；
The described digital information needing identification is carried out being reduced into analog voice signal, the analog voice letter after described reduction
Number it is the final voice needing identification.
In the present embodiment, the digitized process of described voice is voice signal to be digitized sample.
Wherein said eigenvalue includes the amplitude of sound, frequency.
Present invention disclosed above preferred embodiment is only intended to help illustrate the present invention.Preferred embodiment is not detailed
Describe all of details, also do not limit the specific embodiment that this invention is only described.Obviously, the content according to this specification,
Can make many modifications and variations.This specification is chosen and is specifically described these embodiments, is to preferably explain the present invention
Principle and practical application so that skilled artisan can be best understood by and utilize the present invention.The present invention is only
Limited by claims and its four corner and equivalent.
Claims (3)
1. a kind of audio recognition method is it is characterised in that comprise the following steps：
Identified analog voice signal is digitized, becomes and turn to the digital information being suitable to computer disposal；
Described digital information is carried out with pretreatment, removes the interference information being mixed into and reduce deformation or distortion；
Pretreated digital information is carried out with feature selection, selects to need the eigenvalue of the voice of identification, after the pre-treatment
The information comprising described eigenvalue is extracted in digital information；
The eigenvalue needing identification is mapped to feature space so that each needs to identify from object space by characteristics extraction process
One of eigenvalue available feature space point represent；
By the described digital information by retaining the eigenvalue needing identification in pretreated digital information, by other numeral letters
Breath rejects the digital information obtaining needs identification；
The described digital information needing identification is carried out being reduced into analog voice signal, the analog voice signal after described reduction is
For the final voice needing identification.
2. audio recognition method as claimed in claim 1 is it is characterised in that the digitized process of described voice is that voice is believed
Number it is digitized sampling.
3. audio recognition method as claimed in claim 1 is it is characterised in that described eigenvalue includes the amplitude of sound, frequency.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610906925.1A CN106409297A (en) | 2016-10-18 | 2016-10-18 | Voice recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610906925.1A CN106409297A (en) | 2016-10-18 | 2016-10-18 | Voice recognition method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106409297A true CN106409297A (en) | 2017-02-15 |
Family
ID=58012347
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610906925.1A Pending CN106409297A (en) | 2016-10-18 | 2016-10-18 | Voice recognition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106409297A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1104925A1 (en) * | 1999-12-03 | 2001-06-06 | Siemens Aktiengesellschaft | Method for processing speech signals by substracting a noise function |
| US20080228478A1 (en) * | 2005-06-15 | 2008-09-18 | Qnx Software Systems (Wavemakers), Inc. | Targeted speech |
| CN101807398A (en) * | 2009-02-16 | 2010-08-18 | 宏正自动科技股份有限公司 | Speech recognition device and operating method thereof |
| CN103761974A (en) * | 2014-01-28 | 2014-04-30 | 上海力声特医学科技有限公司 | Cochlear implant |
| CN104810024A (en) * | 2014-01-28 | 2015-07-29 | 上海力声特医学科技有限公司 | Double-path microphone speech noise reduction treatment method and system |
-
2016
- 2016-10-18 CN CN201610906925.1A patent/CN106409297A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1104925A1 (en) * | 1999-12-03 | 2001-06-06 | Siemens Aktiengesellschaft | Method for processing speech signals by substracting a noise function |
| US20080228478A1 (en) * | 2005-06-15 | 2008-09-18 | Qnx Software Systems (Wavemakers), Inc. | Targeted speech |
| CN101807398A (en) * | 2009-02-16 | 2010-08-18 | 宏正自动科技股份有限公司 | Speech recognition device and operating method thereof |
| CN103761974A (en) * | 2014-01-28 | 2014-04-30 | 上海力声特医学科技有限公司 | Cochlear implant |
| CN104810024A (en) * | 2014-01-28 | 2015-07-29 | 上海力声特医学科技有限公司 | Double-path microphone speech noise reduction treatment method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108305615B (en) | Object identification method and device, storage medium and terminal thereof | |
| CN110473566A (en) | Audio separation method, device, electronic equipment and computer readable storage medium | |
| CN108831440A (en) | A kind of vocal print noise-reduction method and system based on machine learning and deep learning | |
| WO2010148141A3 (en) | Apparatus and method for speech analysis | |
| CN101887722A (en) | Rapid voiceprint authentication method | |
| CN107274911A (en) | A kind of similarity analysis method based on sound characteristic | |
| CN111081223B (en) | Voice recognition method, device, equipment and storage medium | |
| CN107705791A (en) | Caller identity confirmation method, device and Voiceprint Recognition System based on Application on Voiceprint Recognition | |
| CN106782521A (en) | A kind of speech recognition system | |
| CN105227966A (en) | To televise control method, server and control system of televising | |
| CN110782915A (en) | Waveform music component separation method based on deep learning | |
| CN112349267A (en) | Synthesized voice detection method based on attention mechanism characteristics | |
| CN108665901B (en) | Phoneme/syllable extraction method and device | |
| CN106409297A (en) | Voice recognition method | |
| CN106228984A (en) | Voice recognition information acquisition methods | |
| CN114093369B (en) | A method, device, electronic device and storage medium for separating talkers | |
| CN106485280A (en) | A kind of computer based image-recognizing method | |
| CN106653040A (en) | Voice audio signal sampling processing method | |
| Dutta | Text dependent speaker identification based on spectrograms | |
| CN110265049A (en) | A kind of audio recognition method and speech recognition system | |
| CN104318931A (en) | Emotional activity obtaining method and apparatus of audio file, and classification method and apparatus of audio file | |
| Khonglah et al. | Indoor/Outdoor Audio Classification Using Foreground Speech Segmentation. | |
| Tahliramani et al. | Performance analysis of speaker identification system with and without spoofing attack of voice conversion | |
| CN102592592A (en) | Voice data extraction method and device | |
| CN113571054B (en) | Speech recognition signal preprocessing method, device, equipment and computer storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into 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: 20170215 |