CN103813251B - Hearing-aid denoising device and method allowable for adjusting denoising degree - Google Patents

Hearing-aid denoising device and method allowable for adjusting denoising degree Download PDF

Info

Publication number
CN103813251B
CN103813251B CN201410074153.0A CN201410074153A CN103813251B CN 103813251 B CN103813251 B CN 103813251B CN 201410074153 A CN201410074153 A CN 201410074153A CN 103813251 B CN103813251 B CN 103813251B
Authority
CN
China
Prior art keywords
frame
module
denoising
threshold
wiener filtering
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.)
Active
Application number
CN201410074153.0A
Other languages
Chinese (zh)
Other versions
CN103813251A (en
Inventor
薛风杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen micro nano perception Computing Technology Co., Ltd.
Original Assignee
Shenzhen Micro & Nano Integrated Circuit And System Application Institute
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenzhen Micro & Nano Integrated Circuit And System Application Institute filed Critical Shenzhen Micro & Nano Integrated Circuit And System Application Institute
Priority to CN201410074153.0A priority Critical patent/CN103813251B/en
Publication of CN103813251A publication Critical patent/CN103813251A/en
Application granted granted Critical
Publication of CN103813251B publication Critical patent/CN103813251B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The invention provides a hearing-aid denoising device and method allowable for adjusting denoising degree. The hearing-aid denoising device allowable for adjusting denoising degree comprises an spectrum estimation module, a spectrum mean value module, an endpoint detection module, a Wiener filtering module, a denoising degree setting module and an inverse fast fourier transform module, wherein the endpoint detection module is used for determining whether a current frame is a voice frame, a sub-voice frame, a sub-noise frame or a noise frame and correspondingly setting marks; the Wiener filtering module is used for performing Wiener filtering operation and obtaining Wiener filtering coefficient; the denoising degree setting module is used for determining noise reduction degree according to the marks of the endpoint detection module and the Wiener filtering coefficient of the Wiener filtering module; the inverse fast fourier transform module is used for converting a frequency domain of a processed signal to a time domain so as to obtain a denoised signal. By means of the hearing-aid denoising device and method allowable for adjusting denoising degree, a user can control noise strength suppression on the basis of independent choice, and thereby, operating flexibility and comfortable experience levels of the user can be improved.

Description

A kind of sonifer denoising apparatus and method of scalable denoising degree
Technical field
The present invention relates to the denoising apparatus and method of a kind of sonifer, relate more specifically to the sonifer denoising apparatus and method of a kind of scalable denoising degree.
Background technology
Along with the development of modern science and technology, digital deaf-aid is gradually accepted by increasing Deaf and Hard of Hearing Talents with its powerful signal handling capacity.But, in a noisy environment, hearing aid wearer can degradation to the intelligibility of voice.Therefore, squelch circuit module is particularly important for the speech processes of sonifer.Some the most conventional denoising methods include the methods such as spectrum-subtraction, Wiener Filter Method, subspace sound enhancement method, and most common of which is Wiener Filter Method, and it can remove the background noise in environment, white noise and some music noises etc..
Conventional a kind of Wiener filtering algorithm is ETSI(ETSI) standard ETSI ES 202 050 V1.1.5(2007-01).Fig. 1 is the FB(flow block) of the two-stage Wiener filtering method of prior art.As it is shown in figure 1, Wiener filtering coefficient is transformed into the Mel territory relevant to speech perception by two-stage Mel warpage Wiener filtering application Mel territory triangular filter group, then signal is filtered.
In the first stage, the spectrum of incoming frame is calculated by Power estimation module, spectrum mean module power spectrum between frame front and back is all worth to the power spectrum of time smoothing, and end-point detection (Voice Activity Detection, VADNest) module judges that present frame is speech frame or pure noise frame.After Wiener filtering module completes the calculating of linear frequency filter coefficient, use and beautify band module and carry out smooth operation and obtain Mel warpage Wiener filtering coefficient, then carry out Mel IDCT operation and obtain the time domain impulse response of Mel warpage Wiener filter.Then, filtration module filter impulse response and input speech signal are carried out convolution, thus realize Wiener filtering process.
Carrying out Power estimation when, need to carry out 256 FFT, then carry out frequency domain square obtaining frequency spectrum.Fig. 2 is the FB(flow block) of the end-point detection (VADNest) in the two-stage Wiener filtering method of prior art, and the formula wherein seeking FRAME_EN is:
As it is shown in figure 1, after the Wiener filtering of the second level asks for Mel yardstick each sub-band filter coefficient unlike the first order in beautifying band module, coefficient has been carried out gain process in gain regulation module.The signal frame that signal to noise ratio (snr) is relatively low is utilized gain process make the more degree of depth noise eliminate, to signal frame higher for SNR then by reduce filter factor gain to lower filters affect, thus reduce noise elimination the degree of depth.By such process, reduce further the amplitude of noise signal, retain voice signal as far as possible simultaneously, be conducive to improving the accuracy rate identified.Additionally, go DC Module for eliminating DC component.
From Fig. 1 and Fig. 2 it can be seen that existing algorithm is all directly to judge that current signal frame is speech frame or noise frame, then different frames is carried out different disposal.But, the patient being applied to there is a need to sonifer the most tough is comprehended at direct place, and this is not suitable to everyone.Such as: some people thinks that number voice is not noise, but some people just feels it is noise.Thus result in the biggest trouble.
Summary of the invention
The technical problem to be solved is for different hearing loss patients, and patient can be oneself to adjust the degree of denoising.
The embodiment of the present invention provides the sonifer denoising device of a kind of scalable denoising degree, including Power estimation module, for by input signal overlap framing and the power spectrum of obtaining incoming frame;Spectrum mean module, is coupled in Power estimation module, obtains the power spectrum of time smoothing for the power spectral density of front and back two frame is carried out averaging operation;Endpoint detection module, is used for judging that present frame is speech frame, secondary speech frame, secondary noise frame or noise frame and correspondingly arranges mark;Wiener filtering module, is coupled in endpoint detection module, is used for carrying out Wiener filtering operation and obtaining Wiener filtering coefficient;Denoising degree arranges module, is coupled in endpoint detection module and Wiener filtering module, for determining the degree to noise attentuation according to the mark from endpoint detection module and the Wiener filtering coefficient from Wiener filtering module;And inverse fast Fourier transform module, it is coupled in denoising degree and module is set, for processed signal is converted back time domain from frequency domain, thus obtain the signal after denoising.
The embodiment of the present invention also provides for the sonifer denoising method of a kind of scalable denoising degree, including: by input signal overlap framing and the power spectrum of obtaining incoming frame;The power spectral density of front and back two frame is carried out averaging operation thus obtains the power spectrum of time smoothing;Judge that present frame is speech frame, secondary speech frame, secondary noise frame or noise frame and correspondingly arranges mark;Carry out Wiener filtering operation and obtain Wiener filtering coefficient;The degree to noise attentuation is determined according to mark and Wiener filtering coefficient;And processed signal is converted back time domain from frequency domain, thus obtain the signal after denoising.
The sonifer denoising apparatus and method of the scalable denoising degree that the present invention provides can be controlled the intensity of suppression noise by user from main separation, thus improve the operating flexibility of user and experience comfort level.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, the accompanying drawing used required in embodiment or description of the prior art will be briefly described below, apparently, accompanying drawing in describing below is only some embodiments of the present invention, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the FB(flow block) of the two-stage Wiener filtering method of prior art.
Fig. 2 is the FB(flow block) of the end-point detection in the two-stage Wiener filtering method of prior art.
Fig. 3 is the structural representation of the sonifer denoising device of the scalable denoising degree that one embodiment of the invention provides.
Fig. 4 is the schematic flow sheet of the sonifer denoising method of the scalable denoising degree that one embodiment of the invention provides.
Detailed description of the invention
In order to make the purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.
Fig. 3 is the structural representation of the sonifer denoising device 300 of the scalable denoising degree that one embodiment of the invention provides.Sonifer denoising device 300 includes that Power estimation module 302, spectrum mean module 304, Wiener filtering module 306, denoising degree arrange module 308, end-point detection (VAD) module 310 and inverse fast Fourier transform (IFFT) module 312.
Frame for by input signal overlap framing, is transformed into frequency domain from time domain by fast Fourier transform (FFT) and obtains the power spectrum of incoming frame by Power estimation module 302.Spectrum mean module 304 obtains the power spectrum of time smoothing for the power spectral density of front and back two frame is carried out averaging operation.In one embodiment, Wiener filtering module 306 can carry out two-stage Wiener filtering operation and obtain Wiener filtering coefficient H.In the Noise Estimation of first order filtering stage, according to the testing result of VAD module 310, non-speech segment is updated.In the Noise Estimation of second level filtering stage, the dependency between voice and noise is utilized to be updated.IFFT module 312 is for converting back time domain by processed signal from frequency domain, thus obtains the signal after denoising.Although the most not describing Power estimation module 302, spectrum mean module 304, more concrete operations of Wiener filtering module 306 and IFFT module 312 in detail, those skilled in the art are it should be appreciated that can use any of appropriate technology with upper module and be implemented in combination in.
The VAD module 310 of the present invention is different from the VAD module (VADNest module as shown in Figure 1) of prior art.In the prior art, VAD module is merely by judging simply that present frame (such as, 20ms Frame) is speech frame or noise frame (such as, by comparing incoming frame energy pvad and threshold value TH) arranges Boolean quantity mark flagVAD.If incoming frame energy pvad is more than threshold value TH, then present frame is judged as speech frame (such as speech, music, information tone etc.), arranges flagVAD=1.On the contrary, if incoming frame energy pvad is not more than threshold value TH, then present frame is judged as noise frame (quiet frame), arranges flagVAD=0.Compared to the single setting of prior art, the threshold value of the VAD module 310 in Fig. 3 arranges the most versatile and flexible.Depending on which the variance of unlike signal frame is in interval, we can set several different threshold value: T1, T2, T3.In one embodiment, it is assumed that the variance of current demand signal frame is D, and threshold value T1 < T2 < T3, then:
As D >=T3, present frame is judged as speech frame, and mark flagVAD is set to 11;
As T2≤D, < during T3, present frame is judged as time speech frame, and mark flagVAD is set to 10;
As T1≤D, < during T2, present frame is judged as time noise frame, and mark flagVAD is set to 01;
As D, < during T1, present frame is judged as noise frame, and mark flagVAD is set to 00;
Compared with prior art, the sonifer denoising device 300 of scalable denoising degree according to embodiments of the present invention includes that denoising degree arranges module 308 the most novelly.Utilizing this module, user can control the intensity of suppression noise from main separation, thus improve the operating flexibility of user and experience comfort level.After being obtained Wiener filtering coefficient H by Wiener filtering module 306, Wiener filtering coefficient H is introduced to denoising degree and arranges module 308.Denoising degree arranges module 308 and determines the degree to noise attentuation according to the mark flagVAD from VAD module 310 and the Wiener filtering coefficient H from Wiener filtering module 306.Such as:
When flagVAD is 11, noise attentuation H_o=β H;
When flagVAD is 10, noise attentuation H_o=0.95 β H;
When flagVAD is 01, noise attentuation H_o=0.87 β H;
When flagVAD is 00, noise attentuation H_o=0.65 β H;
Wherein attenuation parameter β value can be tested to join by oneself manual adjustment and obtain.In a preferred embodiment, we limit the span of attenuation parameter β as 0.9 ~ 1.5.
Advantageously, the present invention program, so that VAD judges the most absolute, is when being between noise frame and speech frame for some to be segmented on the contrary, and such user, from the degree of main regulation denoising, makes user obtain more preferable comfort level.By the improvement to VAD module, situation about being between speech frame and noise frame can be distinguished, and for carrying out different decay in the case of different, user's scalable denoising degree.
Fig. 4 is the schematic flow sheet of the sonifer denoising method 400 of the scalable denoising degree that one embodiment of the invention provides.Below with reference to Fig. 3, Fig. 4 is described.
In step S402: by input signal framing and the power spectrum of obtaining incoming frame.Such as, frame by input signal overlap framing, is transformed into frequency domain from time domain by fast Fourier transform (FFT) and obtains the power spectrum of incoming frame by the Power estimation module 302 in Fig. 3.
In step s 404, carry out averaging operation thus obtain the power spectrum of time smoothing.Such as, the spectrum mean module 304 in Fig. 3 carries out averaging operation to the power spectral density of front and back two frame thus obtains the power spectrum of time smoothing.
In step S406, carry out end-point detection and Wiener filtering operation.Such as, the end-point detection in Fig. 3 (VAD) module 310 can depend on which interval is the variance of unlike signal frame be in and judge that present frame is speech frame, secondary speech frame, secondary noise frame or noise frame.Correspondingly, mark flagVAD can be respectively set to 11,10,01 and 00 in one embodiment.Additionally, the Wiener filtering module 306 in Fig. 3 can carry out two-stage Wiener filtering operation.In the Noise Estimation of first order filtering stage, according to the testing result (such as, mark flagVAD) of VAD module 310, non-speech segment is updated.In the Noise Estimation of second level filtering stage, the dependency between voice and noise is utilized to be updated.Wiener filtering module 306 obtains Wiener filtering coefficient H.
In step S408, determine the degree to noise attentuation according to the result that end-point detection and Wiener filtering operate.Such as, the denoising degree in Fig. 3 arranges module 308 and determines the degree to noise attentuation according to the mark flagVAD from VAD module 310 and the Wiener filtering coefficient H from Wiener filtering module 306.In one embodiment, when flagVAD is 11, noise attentuation H_o=β H;When flagVAD is 10, noise attentuation H_o=0.95 β H;When flagVAD is 01, noise attentuation H_o=0.87 β H;When flagVAD is 00, noise attentuation H_o=0.65 β H.Wherein attenuation parameter β value can be tested to join by oneself manual adjustment and obtain.In a preferred embodiment, we limit the span of attenuation parameter β as 0.9 ~ 1.5.
In step S410, the signal of process is converted back time domain from frequency domain, thus obtains the signal after denoising.Such as, processed signal is converted back time domain from frequency domain by the IFFT module 312 in Fig. 3, thus obtains the signal after denoising.
Advantageously, the sonifer denoising apparatus and method of scalable denoising degree according to embodiments of the present invention can be controlled the intensity of suppression noise by user from main separation, thus improve the operating flexibility of user and experience comfort level.
The foregoing is only presently preferred embodiments of the present invention, not in order to limit the present invention, all any amendment, equivalent and improvement etc. made within the spirit and principles in the present invention, should be included within the scope of the present invention.

Claims (10)

1. a sonifer denoising device for scalable denoising degree, including:
Power estimation module, for by input signal overlap framing and the power spectrum of obtaining incoming frame;
Spectrum mean module, is coupled in described Power estimation module, for carrying out the power spectral density of front and back two frame Averaging operation thus obtain the power spectrum of time smoothing;
Endpoint detection module, is used for judging that present frame is speech frame, secondary speech frame, secondary noise frame or makes an uproar Acoustic frame and mark is correspondingly set;
Wiener filtering module, is coupled in described endpoint detection module, is used for carrying out Wiener filtering operation and asking Go out Wiener filtering coefficient;
Denoising degree arranges module, is coupled in described endpoint detection module and described Wiener filtering module, is used for Filter according to the described mark from described endpoint detection module and the described wiener from described Wiener filtering module Both wave system numbers determine the degree to noise attentuation;And
Inverse fast Fourier transform module, is coupled in described denoising degree and arranges module, for by processed Signal converts back time domain from frequency domain, thus obtains the signal after denoising.
2. sonifer denoising device as claimed in claim 1, it is characterised in that described endpoint detection module Judge that described present frame is speech frame, secondary speech frame, secondary makes an uproar by comparing incoming frame energy and multiple threshold values Sound frame or noise frame and multiple mark is correspondingly set.
3. sonifer denoising device as claimed in claim 2, it is characterised in that the plurality of threshold value includes First threshold, Second Threshold and the 3rd threshold value, described first threshold is less than described Second Threshold and institute State Second Threshold and be less than described 3rd threshold value, wherein when described incoming frame energy is less than described first threshold, Described present frame is judged as noise frame and is the first value by described traffic sign placement;When described incoming frame energy During more than or equal to described first threshold and less than described Second Threshold, described present frame is judged as time making an uproar Sound frame and be the second value by described traffic sign placement;When described incoming frame energy is more than or equal to described second threshold When being worth and be less than described three threshold value, described present frame is judged as time speech frame and is set by described mark It is set to the 3rd value;And when described incoming frame energy is more than or equal to described three threshold value, described present frame It is judged as speech frame and is the 4th value by described traffic sign placement.
4. sonifer denoising device as claimed in claim 3, it is characterised in that be masked as first when described During value, described noise attentuation is equal to attenuation parameter and described Wiener filtering coefficient and the product of 0.65;When described When being masked as the second value, described noise attentuation is equal to attenuation parameter and described Wiener filtering coefficient and 0.87 take advantage of Long-pending;When described be masked as three values time, described noise attentuation is equal to attenuation parameter and described Wiener filtering coefficient With 0.95 product;And when described be masked as four values time, described noise attentuation be equal to attenuation parameter and institute State the product of Wiener filtering coefficient.
5. sonifer denoising device as claimed in claim 4, it is characterised in that taking of described attenuation parameter Value scope is 0.9~1.5.
6. a sonifer denoising method for scalable denoising degree, including:
By input signal overlap framing and the power spectrum of obtaining incoming frame;
The power spectral density of front and back two frame is carried out averaging operation thus obtains the power spectrum of time smoothing;
Judge that present frame is speech frame, secondary speech frame, secondary noise frame or noise frame and correspondingly arranges Mark;
Carry out Wiener filtering operation and obtain Wiener filtering coefficient;
The degree to noise attentuation is determined according to described mark and described Wiener filtering coefficient;And
Processed signal is converted back time domain from frequency domain, thus obtains the signal after denoising.
7. sonifer denoising method as claimed in claim 6, it is characterised in that described judgement present frame is Speech frame, secondary speech frame, secondary noise frame or noise frame and the step of mark is correspondingly set includes:
By compare incoming frame energy and multiple threshold values judge described present frame be speech frame, secondary speech frame, Secondary noise frame or noise frame and multiple mark is correspondingly set.
8. sonifer denoising method as claimed in claim 7, it is characterised in that the plurality of threshold value includes First threshold, Second Threshold and the 3rd threshold value, described first threshold is less than described Second Threshold and institute State Second Threshold and be less than described 3rd threshold value, wherein when described incoming frame energy is less than described first threshold, Described present frame is judged as noise frame and is the first value by described traffic sign placement;When described incoming frame energy During more than or equal to described first threshold and less than described Second Threshold, described present frame is judged as time making an uproar Sound frame and be the second value by described traffic sign placement;When described incoming frame energy is more than or equal to described second threshold When being worth and be less than described three threshold value, described present frame is judged as time speech frame and is set by described mark It is set to the 3rd value;And when described incoming frame energy is more than or equal to described three threshold value, described present frame It is judged as speech frame and is the 4th value by described traffic sign placement.
9. sonifer denoising method as claimed in claim 8, it is characterised in that be masked as first when described During value, described noise attentuation is equal to attenuation parameter and described Wiener filtering coefficient and the product of 0.65;When described When being masked as the second value, described noise attentuation is equal to described attenuation parameter and described Wiener filtering coefficient and 0.87 Product;When described be masked as three values time, described noise attentuation is equal to described attenuation parameter and described wiener Filter factor and the product of 0.95;And when described be masked as four values time, described noise attentuation is equal to described Attenuation parameter and the product of described Wiener filtering coefficient.
10. sonifer denoising method as claimed in claim 9, it is characterised in that described attenuation parameter Span is 0.9~1.5.
CN201410074153.0A 2014-03-03 2014-03-03 Hearing-aid denoising device and method allowable for adjusting denoising degree Active CN103813251B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410074153.0A CN103813251B (en) 2014-03-03 2014-03-03 Hearing-aid denoising device and method allowable for adjusting denoising degree

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410074153.0A CN103813251B (en) 2014-03-03 2014-03-03 Hearing-aid denoising device and method allowable for adjusting denoising degree

Publications (2)

Publication Number Publication Date
CN103813251A CN103813251A (en) 2014-05-21
CN103813251B true CN103813251B (en) 2017-01-11

Family

ID=50709362

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410074153.0A Active CN103813251B (en) 2014-03-03 2014-03-03 Hearing-aid denoising device and method allowable for adjusting denoising degree

Country Status (1)

Country Link
CN (1) CN103813251B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104867499A (en) * 2014-12-26 2015-08-26 深圳市微纳集成电路与系统应用研究院 Frequency-band-divided wiener filtering and de-noising method used for hearing aid and system thereof
CN104602157B (en) * 2015-01-28 2018-11-02 深圳市冠旭电子股份有限公司 A kind of earphone noise-reduction method and device
CN107305774B (en) 2016-04-22 2020-11-03 腾讯科技(深圳)有限公司 Voice detection method and device
CN108053834B (en) * 2017-12-05 2020-02-21 北京声智科技有限公司 Audio data processing method, device, terminal and system
CN109616133B (en) * 2018-09-28 2021-11-30 广州智伴人工智能科技有限公司 Environmental noise removing system
CN112002339B (en) * 2020-07-22 2024-01-26 海尔优家智能科技(北京)有限公司 Speech noise reduction method and device, computer-readable storage medium and electronic device

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6415253B1 (en) * 1998-02-20 2002-07-02 Meta-C Corporation Method and apparatus for enhancing noise-corrupted speech
CN1523573A (en) * 2003-09-12 2004-08-25 中国科学院声学研究所 A multichannel speech enhancement method using postfilter
CN1967659A (en) * 2005-11-14 2007-05-23 北京大学科技开发部 Speech enhancement method applied to deaf-aid

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2398913B (en) * 2003-02-27 2005-08-17 Motorola Inc Noise estimation in speech recognition

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6415253B1 (en) * 1998-02-20 2002-07-02 Meta-C Corporation Method and apparatus for enhancing noise-corrupted speech
CN1523573A (en) * 2003-09-12 2004-08-25 中国科学院声学研究所 A multichannel speech enhancement method using postfilter
CN1967659A (en) * 2005-11-14 2007-05-23 北京大学科技开发部 Speech enhancement method applied to deaf-aid

Also Published As

Publication number Publication date
CN103813251A (en) 2014-05-21

Similar Documents

Publication Publication Date Title
US11056130B2 (en) Speech enhancement method and apparatus, device and storage medium
CN103813251B (en) Hearing-aid denoising device and method allowable for adjusting denoising degree
CN111418010B (en) Multi-microphone noise reduction method and device and terminal equipment
Kokkinakis et al. A channel-selection criterion for suppressing reverberation in cochlear implants
CN113823319B (en) Improved speech intelligibility
US20180268798A1 (en) Two channel headset-based own voice enhancement
JP2003520469A (en) Noise reduction apparatus and method
WO2008101324A1 (en) High-frequency bandwidth extension in the time domain
CN103544961B (en) Audio signal processing method and device
JP5834088B2 (en) Dynamic microphone signal mixer
EP1875466A2 (en) Systems and methods for reducing audio noise
DK3008924T3 (en) METHOD OF SIGNAL PROCESSING IN A HEARING SYSTEM AND HEARING SYSTEM
WO2015085946A1 (en) Voice signal processing method, apparatus and server
Hazrati et al. Tackling the combined effects of reverberation and masking noise using ideal channel selection
AU2022205203A1 (en) A noise reduction method and system
CN104867499A (en) Frequency-band-divided wiener filtering and de-noising method used for hearing aid and system thereof
US20200154202A1 (en) Method and electronic device for managing loudness of audio signal
Itoh et al. Environmental noise reduction based on speech/non-speech identification for hearing aids
CN103824563A (en) Hearing aid denoising device and method based on module multiplexing
KR101295727B1 (en) Apparatus and method for adaptive noise estimation
Jin et al. Multi-channel noise reduction for hands-free voice communication on mobile phones
Bhat et al. Smartphone based real-time super gaussian single microphone speech enhancement to improve intelligibility for hearing aid users using formant information
Eneman et al. Evaluation of signal enhancement algorithms for hearing instruments
CN104703108A (en) Wide-dynamic compression algorithm of digital hearing aid under noise condition
Zhang et al. A robust speech enhancement method based on microphone array

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20180126

Address after: 518000 Guangdong city of Shenzhen province Nanshan District Taoyuan Street Xueyuan Road No. 1001 Nanshan Chi Park C3 building 22 floor

Patentee after: Shenzhen micro nano perception Computing Technology Co., Ltd.

Address before: 518000 Shenzhen International Software Park, Shenzhen, Shenzhen, Guangdong Province, 4, 425

Patentee before: Shenzhen Micro & Nano Integrated Circuit and System Application Institute