CN104811559B - Noise-reduction method, communication means and mobile terminal - Google Patents
Noise-reduction method, communication means and mobile terminal Download PDFInfo
- Publication number
- CN104811559B CN104811559B CN201510223926.1A CN201510223926A CN104811559B CN 104811559 B CN104811559 B CN 104811559B CN 201510223926 A CN201510223926 A CN 201510223926A CN 104811559 B CN104811559 B CN 104811559B
- Authority
- CN
- China
- Prior art keywords
- voice data
- sound
- noise
- reduction
- background
- Prior art date
Links
- 238000004891 communication Methods 0.000 title claims abstract description 20
- 230000003203 everyday Effects 0.000 claims description 7
- 230000014509 gene expression Effects 0.000 claims description 6
- 239000000284 extracts Substances 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 description 8
- 230000001755 vocal Effects 0.000 description 8
- 238000000034 methods Methods 0.000 description 6
- 230000004048 modification Effects 0.000 description 2
- 238000006011 modification reactions Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 280000974386 Formant companies 0.000 description 1
- 206010022000 Influenza Diseases 0.000 description 1
- 210000000867 Larynx Anatomy 0.000 description 1
- 210000004072 Lung Anatomy 0.000 description 1
- 210000003928 Nasal Cavity Anatomy 0.000 description 1
- 281000056986 Peoples Daily companies 0.000 description 1
- 210000002105 Tongue Anatomy 0.000 description 1
- 210000000515 Tooth Anatomy 0.000 description 1
- 210000003484 anatomy Anatomy 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010586 diagrams Methods 0.000 description 1
- 238000005516 engineering processes Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 239000000203 mixtures Substances 0.000 description 1
- 210000000056 organs Anatomy 0.000 description 1
- 230000035479 physiological effects, processes and functions Effects 0.000 description 1
- 230000029058 respiratory gaseous exchange Effects 0.000 description 1
- 239000000126 substances Substances 0.000 description 1
Abstract
Description
Technical field
The present invention relates to a kind of noise-reduction method, communication means and mobile terminals.
Background technique
As social development and the progress of economy, mobile phone have become tool indispensable in people's daily life, hand Machine guarantees that other side can not be by noise effect when call under noisy environment, it will usually carry out noise reduction process.
In the prior art, noise-reduction method is double mic (microphone) noise-reduction methods, and this method can use two microphones, Extra microphone not only will increase hardware cost but also the design of meeting image complete machine mechanism, layout, keep handset structure relatively multiple It is miscellaneous.And the voice data handled using double mic noise-reduction methods would generally be by signal enhanced processing, so that voice signal It is easy distortion, voice result of broadcast is poor.
Summary of the invention
The technical problem to be solved by the present invention is in order to overcome in the prior art mobile phone noise-reduction method keep handset structure complicated, The defect of hardware cost height and effect difference, providing a kind of single microphone of utilization can be realized as noise reduction and the better noise reduction of effect Method, communication means and mobile terminal.
The present invention is to solve above-mentioned technical problem by following technical proposals：
A kind of noise-reduction method is used for mobile terminal, and the mobile terminal includes an at least microphone, it is characterized in that, institute The voiceprint of mobile terminal prestored user is stated, the noise-reduction method includes：
Voice data is received by the microphone, wherein the voice data includes voice data and background sound；
The voice data of the user is obtained from the voice data by the voiceprint.
Vocal print (voiceprint) is the sound wave spectrum for the carrying verbal information that electricity consumption acoustic instrument is shown.Human language Generation is a complicated physiology physical process between Body Languages maincenter and vocal organs, the acoustical generator that people uses in speech Official：Tongue, tooth, larynx, lung, nasal cavity everyone widely different in terms of size and form, so the vocal print of any two people Map is all variant.
Application on Voiceprint Recognition (Voiceprint Recognition) has been applied in every field, including identity validation, Criminal investigation work etc., Application on Voiceprint Recognition include that speaker recognizes (Speaker Identification), and in criminal investigation field, speaker is distinguished Recognizing can accomplish to judge in more people's conversational speech that certain section of dialogue is described in which.
The present invention is extracted from one section of voice data comprising voice data and background sound using sound groove recognition technology in e Voice data extracts, and excludes background sound, and the voice data of acquisition is clear, and distortion level is low.
Noise-reduction method of the invention has a wide range of application, including voice data is used to make a phone call, sends voice SMS, society The language of software is handed over to chat and voice memo etc..
Preferably, the mobile terminal includes a display screen, include before receiving voice data：
A text information is shown on the display screen, and receives the voice data that user reads aloud the text information；
From identification set of the extraction sound characteristic to be formed in the voiceprint in the voice data of the text information；
Wherein, the mobile terminal passes through the voice data in the identification set identification voice data, the text envelope Breath includes dialogue everyday expressions.
The task of feature extraction is to extract and select have the characteristics such as separability is strong, stability is high to the vocal print of speaker Acoustics or language feature.Different from speech recognition, the feature of Application on Voiceprint Recognition must be " personalization " feature, and Speaker Identification Feature must be " common feature " for speaker.Although major part Voiceprint Recognition System is all acoustics level at present Feature, but the feature for characterizing a personal touch should be multifaceted, including：With the anatomical structure of the pronunciation mechanism of the mankind Related acoustic feature such as frequency spectrum, cepstrum, formant, fundamental tone, reflection coefficient etc., nasal sound, band deep breathing sound, hoarse sound, Laugh etc..From the angle that can be modeled using mathematical method, feature packet that vocal print automatic identification model can be used at present It includes：Acoustic feature, lexical characteristics, prosodic features, languages, dialect and accent information, channel information etc..
Dialogue everyday expressions may include number, greet the every-day languages frequencies of occurrences such as term, doubt statement, case statement Higher word or sentence.
Preferably, the voiceprint further includes a background set,
The sound characteristic of the background sound of common call scene is preset in the background set；
And/or
The mobile terminal receives voice data, and extracts the sound characteristic of background sound from the voice data with shape At the background set；
The mobile terminal identifies the voice data in voice data by the identification set and background set.
When identification, which is integrated into, can not identify voice data in identification, background set identification background sound can use simultaneously To obtain voice data after background sound is eliminated.Common call scene includes the ambient sound of the ambient sound at station, office Sound and the ambient sound of school etc..
Using the present invention, user can acquire sound in operating position, interchange of position with self-setting background set, user Sound characteristic in the voice data is set as using after background set is used for by data.
Preferably, the voice data is divided into several sound sections according to preset duration, the noise-reduction method includes：
Judge whether voice data is obtained by identification set in current sound section, if then utilizing the drop The next sound section of method for de-noising processing, if otherwise by the background sound in background set removal current sound section with It obtains the voice data and handles next sound section using the noise-reduction method.
Next sound section refers to that processing obtains next sound section of the sound section of voice data recently, When handling " next sound section ", " next sound section " is considered as current sound section, at the noise-reduction method Reason, which refers to, to be judged whether to obtain voice data by identification set in next sound section and be executed according to judging result Follow-up process.The present invention is handled as unit of single sound section, the sound section handled for one, first with Identification set removes background sound using background set if it can not identify to obtain voice data to obtain voice number According to according to the next sound section of timing sequence process until voice data all complete by processing after the completion of a sound section processing.
Preferably, statistics obtains the number of the sound section of voice data by the background set, if the number is big Then pass through double mic noise-reduction methods in one first preset value and handles the voice data in remaining sound section to obtain voice number According to.
If the number is greater than, the first preset value is likely to occur user's flu sound change or other people use the mobile phone, At this moment it will receive influence using the effect of Application on Voiceprint Recognition, thus it is possible to vary improve noise reduction effect at using double mic noise-reduction methods.
Preferably, the voice data is divided into several sound sections according to preset duration, the noise-reduction method includes：
Judge whether voice data is obtained by identification set in current sound section, if then utilizing the drop The next sound section of method for de-noising processing, if otherwise handling the voice data in current sound section by double mic noise-reduction methods To obtain voice data and handle next sound section using the noise-reduction method.
If made a phone call in strange background environment, using background set, the effect is unsatisfactory, thus it is possible to vary Cheng Li Noise reduction effect is improved with double mic noise-reduction methods.
Preferably, the noise-reduction method includes：
The number that the unused identification set obtains the sound section of voice data is counted, if the number is greater than one second Preset value then verifies the identity of the user after the completion of receiving voice data, if the non-rule starting of the identity is described mobile whole The anti-theft modes at end.
Application on Voiceprint Recognition itself can be used as the means of verifying identity, and the application judges that mobile terminal is using Application on Voiceprint Recognition It is no to fall into his manpower the safety for improving mobile phone.Anti-theft modes can mobile phone can not unlock, anti-brush machine mode opens for locking Dynamic, transmission location information etc..
The present invention also provides a kind of communication means, are used for mobile terminal, it is characterized in that, the communication means includes：
In telephone relation or when sending voice messaging, judge in the voice data received whether include legitimate user language Then sound sends the voice data if being then voice data by voice data translation using noise-reduction method as described above.
The mode for sending the voice data include make a phone call, send voice SMS, social software language chat and language Sound memorandum etc..
The present invention provides a kind of mobile terminal again, and the mobile terminal includes an at least microphone and a display screen, Feature is that the voiceprint of the mobile terminal prestored user, the mobile terminal further includes a receiving module, an acquisition mould Block and a processing module,
The display screen is for showing a text information；
The receiving module is used to receive voice data by microphone, and the voice data includes that user reads aloud the text The voice data of this information；
The processing module is used to extract sound characteristic from the voice data of the text information to form the vocal print Identification set in information, and the sound characteristic for extracting background sound from voice data is to form the background set；
The module that obtains is used to obtain the voice data in voice data by the identification set and background set；
Wherein, the text information includes dialogue everyday expressions, and the back of common call scene is preset in the background set The sound characteristic of scape sound.
Preferably, the mobile terminal further includes a judgment module and a statistical module,
The processing module is also used to the voice data being divided into several sound sections according to preset duration；
The judgment module gathers whether obtain voice data by the identification in current sound section for judging, If then handling next sound section using the noise-reduction method, if otherwise removing current sound area by the background set Interior background sound is to obtain the voice data；
The statistical module is used to count the number for the sound section that voice data is obtained by the background set, if institute It states number and passes through the voice data in double mic noise-reduction method processing current sound section then greater than one first preset value to obtain language Sound data.
On the basis of common knowledge of the art, above-mentioned each optimum condition, can any combination to get each preferable reality of the present invention Example.
The positive effect of the present invention is that：Noise-reduction method, communication means and mobile terminal of the invention can be realized Noise when mobile phone communication is reduced using single microphone, effectively save cost simultaneously simplifies mobile phone hardware structure.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of the mobile phone of the embodiment of the present invention 1.
Fig. 2 is the flow chart of the noise-reduction method of the embodiment of the present invention 1.
Fig. 3 is the flow chart of the noise-reduction method of the embodiment of the present invention 3.
Specific embodiment
The present invention is further illustrated below by the mode of embodiment, but does not therefore limit the present invention to the reality It applies among a range.
Embodiment 1
Referring to Fig. 1, the present embodiment provides a mobile phone 1, the mobile phone includes a microphone 11 and a display screen 12, The voiceprint of the mobile phone prestored user, the mobile phone further include that a receiving module 13, one obtains module 14, a judgment module 15, a statistical module 16 and a processing module 17.
The display screen is for showing a text information, wherein the text information includes dialogue everyday expressions.
The receiving module is used to receive voice data by microphone, and the voice data includes that user reads aloud the text The voice data of this information.
The processing module is used to extract sound characteristic from the voice data of the text information to form the vocal print Identification set in information, and the sound characteristic for extracting background sound from voice data is to form the background set.
The module that obtains is used to obtain the voice data in voice data by the identification set and background set.
The sound characteristic of the background sound of common call scene is preset in the background set.
The sound characteristic of common call scene is preset in the background set of the present embodiment, in addition user can also be voluntarily Acquire the sound characteristic of background sound.
The processing module is also used to the voice data being divided into several sound sections according to preset duration (two seconds).
The judgment module gathers whether obtain voice data by the identification in current sound section for judging, If then handling next sound section using the noise-reduction method, if otherwise removing current sound area by the background set Interior background sound is to obtain the voice data；
The statistical module is used to count the number for the sound section that voice data is obtained by the background set, if institute The identity that number verifies the user greater than 50 after the completion of receiving voice data is stated, if described in the non-rule starting of the identity The anti-theft modes of mobile phone.
Referring to fig. 2, using above-mentioned mobile phone, the present embodiment also provides a kind of noise-reduction method, and the noise-reduction method includes：
Step 100 shows a text information on the display screen, and by the microphone receive user read aloud it is described The voice data of text information.
The voice data includes voice data and background sound.
Step 101, from the voice data of the text information extract sound characteristic to be formed in the voiceprint Identification set, and the sound characteristic of background sound is extracted from the voice data to form the background set of voiceprint.
The mobile phone is first carrying out Initialize installation, that is, the voiceprint and background sound of acquisition user using preceding Sound characteristic, the acquisition of sound characteristic in the acquisition and background set of the sound characteristic in identification set can be separated It carries out, for example, when reading aloud the text information, only acquisition identifies the sound characteristic in set to user, user will after the completion of acquisition Mobile phone takes job site to acquire the sound characteristic in background set, and the present embodiment to acquire identification set and background collection simultaneously For the sound characteristic of conjunction, achieve the purpose that quickly to form voiceprint.
The sound characteristic of the background sound of common call scene is preset in the background set of the present embodiment.
Step 102 dials and carries out mobile phone communication, then receives voice data of the user for call.
The voice data is divided into several sound sections according to preset duration (two seconds) by step 103, utilizes the knowledge The voice data in set identification voice data Ji He not obtained.
Step 104 judges whether obtain voice data by identification set in current sound section, if then holding Row step 106, thens follow the steps 105 if not.
Step 105 removes the background sound in current sound section by the background set to obtain the voice number According to.
Step 106, judge it is nearest one acquisition voice data sound section whether be the voice data last A sound section, if then terminating process, if otherwise obtaining nearest one next sound of the sound section of voice data Section is as current sound section and then return step 104.
The noise-reduction method further includes：
Count it is unused it is described identification set obtain voice data sound section number, if the number be greater than 50 if The identity that voice data verifies the user after the completion is received, if the non-rule of the identity starts the anti-theft modes of the mobile phone.
Using above-mentioned noise-reduction method, the present embodiment also provides a kind of communication means, and the communication means includes：
Step 102 include in the voice data that receives of judgement whether include legitimate user voice, if then by voice number According to the receiving end for being sent to call.Concrete mode is that every processing acquisition voice data for completing a sound section just sends one It is a.
The noise-reduction method, communication means and mobile phone of the present embodiment can be realized when reducing mobile phone communication using single microphone Noise, effectively save cost simultaneously simplify mobile phone hardware structure.
Embodiment 2
The present embodiment is substantially the same manner as Example 1, the difference is that only：
The mobile phone includes two microphones.
The noise-reduction method includes：
Statistics obtains the number of the sound section of voice data by the background set, leads to if the number is greater than 10 It crosses double mic noise-reduction methods and handles the voice data in remaining sound section to obtain voice data.
The noise-reduction method of the present embodiment realizes noise reduction using two ways, further ensures that speech quality.
Embodiment 3
Referring to Fig. 3, the present embodiment is substantially the same manner as Example 1, the difference is that only that step 105 replaces with：
Step 105 handles the voice data in current sound section by double mic noise-reduction methods to obtain voice data.
The noise-reduction method, communication means and mobile phone of the present embodiment can be realized when reducing mobile phone communication using single microphone Noise, effectively save cost simultaneously simplify mobile phone hardware structure.
Although specific embodiments of the present invention have been described above, it will be appreciated by those of skill in the art that these It is merely illustrative of, protection scope of the present invention is defined by the appended claims.Those skilled in the art is not carrying on the back Under the premise of from the principle and substance of the present invention, many changes and modifications may be made, but these are changed Protection scope of the present invention is each fallen with modification.
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510223926.1A CN104811559B (en) | 2015-05-05 | 2015-05-05 | Noise-reduction method, communication means and mobile terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510223926.1A CN104811559B (en) | 2015-05-05 | 2015-05-05 | Noise-reduction method, communication means and mobile terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104811559A CN104811559A (en) | 2015-07-29 |
| CN104811559B true CN104811559B (en) | 2018-11-20 |
Family
ID=53696045
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510223926.1A CN104811559B (en) | 2015-05-05 | 2015-05-05 | Noise-reduction method, communication means and mobile terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104811559B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106373577A (en) * | 2016-08-18 | 2017-02-01 | 胡伟 | Personal voice system |
| CN106487410B (en) * | 2016-12-13 | 2019-06-04 | 北京奇虎科技有限公司 | A kind of authority control method and device of message interruption-free |
| CN106791122A (en) * | 2016-12-27 | 2017-05-31 | 广东小天才科技有限公司 | The call control method and wearable device of a kind of wearable device |
| CN106920559B (en) * | 2017-03-02 | 2020-10-30 | 奇酷互联网络科技（深圳）有限公司 | Voice communication optimization method and device and call terminal |
| CN107393540A (en) * | 2017-07-20 | 2017-11-24 | 任文 | A kind of method that phonetic entry abates the noise |
| CN107369441A (en) * | 2017-09-08 | 2017-11-21 | 奇酷互联网络科技（深圳）有限公司 | Noise-eliminating method, device and the terminal of voice signal |
| CN109065066B (en) * | 2018-09-29 | 2020-03-31 | 广东小天才科技有限公司 | Call control method, device and equipment |
| CN109272996A (en) * | 2018-11-09 | 2019-01-25 | 广州长嘉电子有限公司 | A kind of noise-reduction method and system |
| CN109817196A (en) * | 2019-01-11 | 2019-05-28 | 安克创新科技股份有限公司 | A kind of method of canceling noise, device, system, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101668085A (en) * | 2009-09-16 | 2010-03-10 | 宇龙计算机通信科技(深圳)有限公司 | Method for regulating voice output of mobile terminal and mobile terminal |
| CN102694891A (en) * | 2011-03-21 | 2012-09-26 | 鸿富锦精密工业（深圳）有限公司 | System and method for removing conversation noises |
| CN103456305A (en) * | 2013-09-16 | 2013-12-18 | 东莞宇龙通信科技有限公司 | Terminal and speech processing method based on multiple sound collecting units |
| CN103514884A (en) * | 2012-06-26 | 2014-01-15 | 华为终端有限公司 | Communication voice denoising method and terminal |
| CN103971696A (en) * | 2013-01-30 | 2014-08-06 | 华为终端有限公司 | Method, device and terminal equipment for processing voice |
-
2015
- 2015-05-05 CN CN201510223926.1A patent/CN104811559B/en active IP Right Grant
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101668085A (en) * | 2009-09-16 | 2010-03-10 | 宇龙计算机通信科技(深圳)有限公司 | Method for regulating voice output of mobile terminal and mobile terminal |
| CN102694891A (en) * | 2011-03-21 | 2012-09-26 | 鸿富锦精密工业（深圳）有限公司 | System and method for removing conversation noises |
| CN103514884A (en) * | 2012-06-26 | 2014-01-15 | 华为终端有限公司 | Communication voice denoising method and terminal |
| CN103971696A (en) * | 2013-01-30 | 2014-08-06 | 华为终端有限公司 | Method, device and terminal equipment for processing voice |
| CN103456305A (en) * | 2013-09-16 | 2013-12-18 | 东莞宇龙通信科技有限公司 | Terminal and speech processing method based on multiple sound collecting units |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104811559A (en) | 2015-07-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105009204B (en) | Speech recognition power management | |
| US9769296B2 (en) | Techniques for voice controlling bluetooth headset | |
| Nakamura et al. | Speaking-aid systems using GMM-based voice conversion for electrolaryngeal speech | |
| US8825479B2 (en) | System and method for recognizing emotional state from a speech signal | |
| CN105096940B (en) | Method and apparatus for carrying out speech recognition | |
| US9552815B2 (en) | Speech understanding method and system | |
| EP2783365B1 (en) | Method and system for adapting grammars in hybrid speech recognition engines for enhancing local speech recognition performance | |
| WO2017054122A1 (en) | Speech recognition system and method, client device and cloud server | |
| US8706488B2 (en) | Methods and apparatus for formant-based voice synthesis | |
| JP4546512B2 (en) | Speech recognition system using technology that implicitly adapts to the speaker | |
| Campbell et al. | Forensic speaker recognition | |
| Rabiner et al. | Introduction to digital speech processing | |
| EP1199708B1 (en) | Noise robust pattern recognition | |
| US7676372B1 (en) | Prosthetic hearing device that transforms a detected speech into a speech of a speech form assistive in understanding the semantic meaning in the detected speech | |
| US8244540B2 (en) | System and method for providing a textual representation of an audio message to a mobile device | |
| CN102723078B (en) | Emotion speech recognition method based on natural language comprehension | |
| US8660842B2 (en) | Enhancing speech recognition using visual information | |
| KR20180091903A (en) | METHOD, APPARATUS AND STORAGE MEDIUM FOR CONFIGURING VOICE DECODING NETWORK IN NUMERIC VIDEO RECOGNI | |
| CN105161093B (en) | A kind of method and system judging speaker's number | |
| US7082395B2 (en) | Signal injection coupling into the human vocal tract for robust audible and inaudible voice recognition | |
| JP4796309B2 (en) | Method and apparatus for multi-sensor speech improvement on mobile devices | |
| CN102388416B (en) | Signal processing apparatus and signal processing method | |
| WO2015161240A2 (en) | Speaker verification | |
| US7684982B2 (en) | Noise reduction and audio-visual speech activity detection | |
| CN104575504A (en) | Method for personalized television voice wake-up by voiceprint and voice identification |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |