CN113035222A - Voice noise reduction method and device, filter determination method and voice interaction equipment - Google Patents
Voice noise reduction method and device, filter determination method and voice interaction equipment Download PDFInfo
- Publication number
- CN113035222A CN113035222A CN202110219713.7A CN202110219713A CN113035222A CN 113035222 A CN113035222 A CN 113035222A CN 202110219713 A CN202110219713 A CN 202110219713A CN 113035222 A CN113035222 A CN 113035222A
- Authority
- CN
- China
- Prior art keywords
- signal
- sound
- sound signal
- noise
- voice
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 230000009467 reduction Effects 0.000 title claims abstract description 31
- 230000003993 interaction Effects 0.000 title claims abstract description 11
- 238000001914 filtration Methods 0.000 claims abstract description 48
- 230000005236 sound signal Effects 0.000 claims description 118
- 238000001228 spectrum Methods 0.000 claims description 69
- 230000007613 environmental effect Effects 0.000 claims description 26
- 238000012545 processing Methods 0.000 claims description 20
- 230000000694 effects Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000010408 sweeping Methods 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 1
- 241001417527 Pempheridae Species 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
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
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
- G10L21/0216—Noise filtering characterised by the method used for estimating noise
-
- 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
- G10L15/20—Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech
-
- 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
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
- G10L21/0216—Noise filtering characterised by the method used for estimating noise
- G10L21/0232—Processing in the frequency domain
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Signal Processing (AREA)
- Soundproofing, Sound Blocking, And Sound Damping (AREA)
- Circuit For Audible Band Transducer (AREA)
- Telephone Function (AREA)
Abstract
The application provides a voice noise reduction method and device, a filter determination method and voice interaction equipment. This application can guarantee that smart machine obtains the speech signal that the SNR is higher before carrying out speech recognition through the setting to wave filter filtering parameter to guarantee speech recognition's precision, improve smart machine's work efficiency.
Description
Technical Field
The present application relates to the field of speech signal processing technologies, and in particular, to a method and an apparatus for speech noise reduction, a method for determining a filter, and a speech interaction device.
Background
In recent years, with the development of artificial intelligence, a man-machine interaction mode based on an intelligent voice technology has a profound influence on products such as automobiles, home appliances (floor sweeper, air conditioner and the like), consumer electronics (earphones and intelligent sound boxes) and the like. Under the man-machine interaction mode, the user speaks a voice control instruction to the intelligent equipment, and the intelligent equipment analyzes the voice control instruction to realize corresponding operation, so that the life of human beings is greatly facilitated.
However, in various practical use scenarios, the signal received by the smart device inevitably includes both the voice signal of the user and the ambient noise signal due to the common existence of non-negligible ambient noise. The noise will affect the recognition accuracy of the user speech by the smart device. When the signal received by the smart device has a very low signal-to-noise ratio, the smart device may misrecognize or even fail to recognize the user's voice command at all.
Disclosure of Invention
Aiming at the defects of the prior art, the voice noise reduction method and device, the filter determination method and the voice interaction equipment are provided, the noise signal is offset through the equivalent of the filter, the user voice with higher signal-to-noise ratio can be extracted, and therefore the voice recognition precision is improved. The technical scheme is specifically adopted in the application.
First, to achieve the above object, a method for reducing noise in a speech is provided, which includes: acquiring a sound signal; performing signal processing on the sound signals according to corresponding first filtering parameters, filtering out environmental noise signals in the sound signals, and obtaining voice signals in the sound signals; wherein the first filtering parameter is determined based on a frequency spectrum of the sound signal and an estimated frequency spectrum of the ambient noise signal.
Optionally, the method for reducing noise in voice according to any of the above embodiments, wherein before performing signal processing on the sound signal according to the corresponding first filtering parameter, the method further includes: dividing the sound signal into a plurality of sound signal frames; the signal processing of the sound signal according to the corresponding first filtering parameter comprises: for each sound signal frame, respectively performing the signal processing according to a second filtering parameter corresponding to the sound signal frame, filtering an environmental noise signal in the sound signal frame, and obtaining a voice signal in the sound signal frame; wherein the second filtering parameter is determined based on a frequency spectrum of the sound signal frame and an estimated frequency spectrum of an ambient noise signal in the sound signal frame.
Optionally, the method for reducing noise in speech according to any of the above embodiments, wherein the estimated spectrum of the ambient noise signal in the sound signal frame is determined according to signal-to-noise ratios of a plurality of previous sound signal frames.
Optionally, the method for reducing noise in speech according to any of the above, wherein the first filtering parameter is determined based on a ratio between an estimated magnitude spectrum of the ambient noise signal and a magnitude spectrum of the sound signal.
Optionally, the method for reducing noise in speech according to any of the above, wherein the first filtering parameter is determined based on a ratio between an estimated self-power spectrum of the ambient noise signal and a self-power spectrum of the sound signal.
Optionally, the method for reducing noise in voice according to any of the above embodiments, wherein before performing signal processing on the sound signal according to the corresponding first filtering parameter, the method further includes: and taking the frequency spectrums of the sound signal frames of the sound signal with the preset number as the estimated frequency spectrums of the environment noise signal.
Simultaneously, in order to realize above-mentioned purpose, this application still provides a device of making an uproar falls in pronunciation, and it includes: the signal acquisition module is used for acquiring a sound signal; the voice noise reduction module is used for carrying out signal processing on the sound signals according to the corresponding first filtering parameters, filtering the environmental noise signals in the sound signals and obtaining the voice signals in the sound signals; wherein the first filtering parameter is determined based on a frequency spectrum of the sound signal and an estimated frequency spectrum of the ambient noise signal.
In addition, the present application also provides a computer device, comprising: a processor; a memory comprising computer instructions stored thereon that, when executed by the processor, cause the processor to perform the speech noise reduction method of any of the above.
The application also provides a method for determining a filter, wherein the filter is applied to voice noise reduction, and the filter parameters are determined through the following steps: acquiring a sound signal; estimating an environmental noise component in the sound signal, and determining a noise estimation signal; determining filtering parameters of the filter based on a frequency spectrum of the sound signal and a frequency spectrum of the noise estimation signal.
The application provides a voice interaction device, it includes: the sound acquisition device is used for acquiring sound signals; the filter determined according to the method is connected with the sound acquisition device and used for filtering the environmental noise signal in the sound signal to obtain the voice signal in the sound signal.
Advantageous effects
This application can guarantee that smart machine obtains the speech signal that the SNR is higher before carrying out speech recognition through the setting to wave filter filtering parameter to guarantee speech recognition's precision, improve smart machine's work efficiency.
Additional features and advantages of the application will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the application.
Drawings
The accompanying drawings are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the principles of the application and not limit the application. In the drawings:
FIG. 1 is a flow chart of the steps of the speech noise reduction method of the present application.
Detailed Description
In order to make the purpose and technical solutions of the embodiments of the present application clearer, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings of the embodiments of the present application. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the described embodiments of the application without any inventive step, are within the scope of protection of the application.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
In a first embodiment of the present application, referring to fig. 1, a speech noise reduction method is provided to remove environmental noise from a sound signal collected by a sound collection device, and recover a clean user speech with a higher signal-to-noise ratio from the sound signal. The speech noise reduction method 10 comprises the following steps:
s11, acquiring a sound signal containing environmental noise; and
s14, processing the sound signal according to the first filtering parameter corresponding to the sound signal, filtering the environment noise signal in the sound signal, and obtaining the voice signal in the sound signal;
wherein the first filtering parameter is determined based on a frequency spectrum of the sound signal and an estimated frequency spectrum of the ambient noise signal.
Therefore, when the voice recognition is carried out on the rear end of the intelligent device, the obtained voice signal can improve the accuracy of the voice recognition, so that the working efficiency of the intelligent device is improved.
The sound signal containing the environmental noise can be acquired through a sound signal acquisition device positioned at the front end of intelligent equipment such as a mobile phone, a computer and an intelligent sound box. It can be configured as a single microphone that collects the acoustic signal in the working environment together with the ambient noise and the user's voice:
xm(n)＝xs(n)+xd(n) formula (1)
Wherein x ism(n) is the collected sound signal, xs(n) is xm(n) the clean speech signal, xd(n) is an ambient noise signal.
In some embodiments, before performing the signal processing based on step S14, the method 10 may further include the steps of:
s12, the audio signal is divided into a plurality of audio signal frames.
Therefore, the step S14 of obtaining the speech signal in the sound signal by performing signal processing on the sound signal according to the first filtering parameter corresponding to the sound signal may specifically include, for each sound signal frame, the steps of:
and S142, performing the signal processing process according to the second filtering parameter corresponding to the sound signal frame, filtering the environmental noise signal in the sound signal frame, and obtaining the voice signal in the sound signal frame.
Therefore, the voice signals in each sound signal frame can be respectively obtained, and the required voice can be completely restored by splicing the corresponding frames. Wherein the second filter parameter is determined based on the frequency spectrum of the sound signal frame and the estimated frequency spectrum of the ambient noise signal in the sound signal frame.
In view of the small scale of the frame length, the speech signal and the ambient noise signal contained in each sound signal frame can be regarded as stationary signals, and the statistical characteristics do not change with time, so in some embodiments, the estimated spectrum of the ambient noise signal in the kth sound signal frame can be determined according to the signal-to-noise ratio of a plurality of previous sound signal frames and the spectrum of the kth sound signal frame, such as the (k-1) th sound signal frame, the (k-2) th sound signal frame.
After the voice noise reduction method is applied to intelligent equipment such as mobile phones, computers and intelligent sound boxes, more reliable voice signal input can be provided for a voice recognition system in the intelligent equipment, and therefore more accurate and rapid voice recognition effect is achieved.
In other implementations, the present application may also provide a speech noise reduction apparatus that may be used as a front end of a speech recognition system. This pronunciation noise reduction device includes:
the signal acquisition module is used for acquiring a sound signal mixed with environmental noise and user voice; and
the voice noise reduction module is used for carrying out signal processing on the voice signal according to the corresponding first filtering parameter, filtering an environmental noise signal in the voice signal and obtaining a voice signal in the voice signal; wherein the first filtering parameter is determined based on a frequency spectrum of the sound signal and an estimated frequency spectrum of the ambient noise signal.
Therefore, the voice signal obtained by the voice noise reduction device can be used for Natural Language Processing (NLP) at the back end of the voice recognition system, so that a corresponding instruction is recognized, and the intelligent equipment is controlled to execute corresponding operation according to the instruction information. Because the voice recognition system effectively filters the environmental noise in the voice signal obtained by the voice noise reduction device and has higher signal-to-noise ratio, the interference on the voice signal is smaller when the voice recognition system carries out command recognition, and the correct voice command can be obtained more quickly and accurately to execute corresponding operation.
In summary, the frequency domain expression of equation (1) with respect to step S14 is: sound signal x mixed with environmental noise by Fourier transformm(n) converting to frequency domain to obtain frequency spectrum X of sound signalm(f) In that respect From the frequency spectrum X of a sound signal mixed with ambient noisem(f) Subtracting the noise spectrum X corresponding to the frequency domain feature of the environmental noised(f) Obtaining a speech spectrum Xs(f) To speech audio spectrum Xs(f) Pure speech signal x can be recovered by performing inverse Fourier transforms(n)。
In order to equivalently cancel the ambient noise signal in the sound signal through the filtering-based signal processing process, the filter response h (f) corresponding to the first filtering parameter needs to satisfy:
|Xs(f)|＝|Xm(f) | H (f) | formula (2)
Wherein, | Xm(f)|、|Xs(f) L are respectively the sound signal xm(n) speech signal xs(n) frequency spectrum Xm(f)、Xs(f) Amplitude spectrum of (1).
Thus, step S14 is based on the first filter parameter to extract the sound signal spectrum X mixed with the environmental noise from the sound signal spectrum Xm(f) Subtracting the ambient noise spectrum Xd(f) I.e. the amplitude spectrum | X of the sound signal mixed with the ambient noise by the first filter parameterm(f) I, frequency weighting is carried out, so that an amplitude spectrum | X of the voice signal is obtaineds(f)|。
To further recover the speech signal in the audio signal, the amplitude spectrum | X can be adjusteds(f) I andsound signal xm(n) frequency spectrum Xm(f) Phase spectrum of
Performing inverse Fourier transform to obtain amplitude information and phase information of voice respectively to recover voice signal xs(n)。
In some embodiments, the first filtering parameter may be based on an estimated magnitude spectrum of the ambient noise signal
With the amplitude spectrum | X of the sound signalm(f) The ratio between |, λ (f). Since the environmental noise and the voice generally have no correlation, for the above equation (2), the filter response h (f) corresponding to the first filtering parameter is expressed as:
in still other embodiments, the first filtering parameter may also be based on an estimated self-power spectrum of the ambient noise signal
Self-powered spectrum X with sound signalm(f)·X* m(f)＝|Xm(f)|2The ratio μ (f) therebetween. Since the environmental noise and the voice generally have no correlation, for the above equation (2), the filter response h (f) corresponding to the first filtering parameter is expressed as:
from the above, it can be seen that the key to determining the first filter parameter is to estimate the spectrum of the ambient noise signal a priori. In an actual scene of voice interaction between a user and an intelligent device, the first N frames of a sound signal collected after a microphone is turned on are generally considered to be free of voice, that is, a voice activity event does not occur temporarily in background noise of an environment. Therefore, in some embodiments, before performing the signal processing based on step S14, the method 10 may further include the steps of:
s13, the frequency spectrum of the first preset number of sound signal frames of the sound signal is used as the estimated frequency spectrum of the environmental noise signal.
Specifically, for example, fourier transform may be performed on each of the first N frames of the sound signal to obtain N frequency spectrums, and the average value of the N frequency spectrums is taken as the estimated frequency spectrum of the environmental noise signal.
As mentioned above, the prior estimation of the frequency spectrum of the ambient noise signal is the key to determine the first filter parameter. For an approximately stationary noise environment, the estimated magnitude spectrum of the ambient noise signal determined based on step S13
And the determined first filtering parameters are sufficient for step S14 to achieve effective speech noise reduction;
for a possibly evolving noise environment, it is more preferable that the estimated spectrum of the environmental noise signal in the current sound signal frame is determined based on the signal-to-noise ratios of a plurality of previous sound signal frames, i.e. the estimated amplitude spectrum of the environmental noise signal in the kth sound signal frame is determined as described above
The a posteriori correction may be performed based on the snr of the 1 st to (k-1) th frames of the audio signal. Accordingly, for each sound signal frame, the second filter parameter determined based on the following equation (5) or (6) enables step S142 to implement speech noise reduction within the frame.
Other embodiments of the present application also provide a computer device, comprising: a processor and a memory.
Wherein the memory includes computer instructions stored thereon which, when executed by the processor, cause the processor to perform the speech noise reduction method 10 as provided in any of the embodiments above.
On the other hand, the embodiment of the present application further provides a method for determining a filter, which is used for determining a filtering parameter required by a speech noise reduction scene. The filter determination method 20 comprises the following steps:
and S22, acquiring the sound signal.
S24, the ambient noise component in the sound signal is estimated, and a noise estimation signal is determined.
S26, filter parameters of the filter are determined based on the frequency spectrum of the sound signal and the frequency spectrum of the noise estimation signal.
Under this implementation, this application still provides a voice interaction equipment, for example bluetooth headset, intelligent audio amplifier, acoustic control machine of sweeping the floor etc. it includes:
sound collection means, such as at least one microphone, for collecting sound signals; and
the filter determined according to the method 20 is connected to the sound collecting device, and is configured to filter the environmental noise signal in the sound signal, and obtain the speech signal in the sound signal.
For a Bluetooth earphone product, the obtained voice signal can be sent to a mobile phone end through Bluetooth and then sent to a call far end of an earphone user through a base station; for household appliances such as intelligent sound boxes, voice control floor sweeping machines and the like, the obtained voice signals can be used for voice command recognition by a voice recognition system.
The above are merely embodiments of the present application, and the description is specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the protection scope of the present application.
Claims (10)
1. A method for speech noise reduction, comprising:
acquiring a sound signal;
performing signal processing on the sound signals according to corresponding first filtering parameters, filtering out environmental noise signals in the sound signals, and obtaining voice signals in the sound signals;
wherein the first filtering parameter is determined based on a frequency spectrum of the sound signal and an estimated frequency spectrum of the ambient noise signal.
2. The speech noise reduction method of claim 1, wherein prior to signal processing the sound signal according to the corresponding first filter parameters, the method further comprises: dividing the sound signal into a plurality of sound signal frames;
the signal processing of the sound signal according to the corresponding first filtering parameter comprises: for each sound signal frame, respectively performing the signal processing according to a second filtering parameter corresponding to the sound signal frame, filtering an environmental noise signal in the sound signal frame, and obtaining a voice signal in the sound signal frame; wherein the second filtering parameter is determined based on a frequency spectrum of the sound signal frame and an estimated frequency spectrum of an ambient noise signal in the sound signal frame.
3. The speech noise reduction method of claim 2, wherein the estimated spectrum of the ambient noise signal in the acoustic signal frame is determined based on signal-to-noise ratios of the previous acoustic signal frames.
4. The method of speech noise reduction according to any of claims 1-3, wherein the first filter parameter is determined based on a ratio between an estimated magnitude spectrum of the ambient noise signal and a magnitude spectrum of the sound signal.
5. The speech noise reduction method of any one of claims 1-3, wherein the first filtering parameter is determined based on a ratio between an estimated self-power spectrum of the ambient noise signal and a self-power spectrum of the sound signal.
6. The speech noise reduction method of claim 2, wherein prior to signal processing the sound signal according to the corresponding first filter parameters, the method further comprises: and taking the frequency spectrums of the sound signal frames of the sound signal with the preset number as the estimated frequency spectrums of the environment noise signal.
7. A speech noise reduction apparatus, comprising:
the signal acquisition module is used for acquiring a sound signal;
the voice noise reduction module is used for carrying out signal processing on the sound signals according to the corresponding first filtering parameters, filtering the environmental noise signals in the sound signals and obtaining the voice signals in the sound signals;
wherein the first filtering parameter is based on a frequency spectrum of the sound signal and the ambient noise signal
Estimating the frequency spectrum.
8. A computer device, comprising:
a processor;
a memory including computer instructions stored thereon, the computer instructions being executed
The processor, when executed, causes the processor to perform the speech reduction of any of claims 1-6
And (4) a noise method.
9. A method for determining a filter to be applied to speech noise reduction, comprising:
acquiring a sound signal;
estimating an environmental noise component in the sound signal, and determining a noise estimation signal;
determining filtering parameters of the filter based on a frequency spectrum of the sound signal and a frequency spectrum of the noise estimation signal.
10. A voice interaction device, comprising:
the sound acquisition device is used for acquiring sound signals;
the filter determined according to the method of claim 9, connected to the sound collecting device, for filtering out an ambient noise signal in the sound signal to obtain a speech signal in the sound signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110219713.7A CN113035222B (en) | 2021-02-26 | 2021-02-26 | Voice noise reduction method and device, filter determination method and voice interaction equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110219713.7A CN113035222B (en) | 2021-02-26 | 2021-02-26 | Voice noise reduction method and device, filter determination method and voice interaction equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113035222A true CN113035222A (en) | 2021-06-25 |
| CN113035222B CN113035222B (en) | 2023-10-27 |
Family
ID=76461967
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110219713.7A Active CN113035222B (en) | 2021-02-26 | 2021-02-26 | Voice noise reduction method and device, filter determination method and voice interaction equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113035222B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6032114A (en) * | 1995-02-17 | 2000-02-29 | Sony Corporation | Method and apparatus for noise reduction by filtering based on a maximum signal-to-noise ratio and an estimated noise level |
| US6351731B1 (en) * | 1998-08-21 | 2002-02-26 | Polycom, Inc. | Adaptive filter featuring spectral gain smoothing and variable noise multiplier for noise reduction, and method therefor |
| US20070172073A1 (en) * | 2006-01-26 | 2007-07-26 | Samsung Electronics Co., Ltd. | Apparatus and method of reducing noise by controlling signal to noise ratio-dependent suppression rate |
| US20120095755A1 (en) * | 2009-06-19 | 2012-04-19 | Fujitsu Limited | Audio signal processing system and audio signal processing method |
| GB201505361D0 (en) * | 2015-03-27 | 2015-05-13 | Toshiba Res Europ Ltd | A speech processing device |
| CN106340303A (en) * | 2016-09-20 | 2017-01-18 | 南京朗逸锐科电子科技有限公司 | Speech denoising method based on time frequency domain |
| CN108391190A (en) * | 2018-01-30 | 2018-08-10 | 努比亚技术有限公司 | A kind of noise-reduction method, earphone and computer readable storage medium |
| CN108735229A (en) * | 2018-06-12 | 2018-11-02 | 华南理工大学 | A kind of amplitude based on noise Ratio Weighted and phase combining compensation anti-noise sound enhancement method and realization device |
| CN108831500A (en) * | 2018-05-29 | 2018-11-16 | 平安科技（深圳）有限公司 | Sound enhancement method, device, computer equipment and storage medium |
| CN111292758A (en) * | 2019-03-12 | 2020-06-16 | 展讯通信（上海）有限公司 | Voice activity detection method and device and readable storage medium |
-
2021
- 2021-02-26 CN CN202110219713.7A patent/CN113035222B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6032114A (en) * | 1995-02-17 | 2000-02-29 | Sony Corporation | Method and apparatus for noise reduction by filtering based on a maximum signal-to-noise ratio and an estimated noise level |
| US6351731B1 (en) * | 1998-08-21 | 2002-02-26 | Polycom, Inc. | Adaptive filter featuring spectral gain smoothing and variable noise multiplier for noise reduction, and method therefor |
| US20070172073A1 (en) * | 2006-01-26 | 2007-07-26 | Samsung Electronics Co., Ltd. | Apparatus and method of reducing noise by controlling signal to noise ratio-dependent suppression rate |
| US20120095755A1 (en) * | 2009-06-19 | 2012-04-19 | Fujitsu Limited | Audio signal processing system and audio signal processing method |
| GB201505361D0 (en) * | 2015-03-27 | 2015-05-13 | Toshiba Res Europ Ltd | A speech processing device |
| CN106340303A (en) * | 2016-09-20 | 2017-01-18 | 南京朗逸锐科电子科技有限公司 | Speech denoising method based on time frequency domain |
| CN108391190A (en) * | 2018-01-30 | 2018-08-10 | 努比亚技术有限公司 | A kind of noise-reduction method, earphone and computer readable storage medium |
| CN108831500A (en) * | 2018-05-29 | 2018-11-16 | 平安科技（深圳）有限公司 | Sound enhancement method, device, computer equipment and storage medium |
| CN108735229A (en) * | 2018-06-12 | 2018-11-02 | 华南理工大学 | A kind of amplitude based on noise Ratio Weighted and phase combining compensation anti-noise sound enhancement method and realization device |
| CN111292758A (en) * | 2019-03-12 | 2020-06-16 | 展讯通信（上海）有限公司 | Voice activity detection method and device and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113035222B (en) | 2023-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109727604B (en) | Frequency domain echo cancellation method for speech recognition front end and computer storage medium | |
| CN109767783B (en) | Voice enhancement method, device, equipment and storage medium | |
| CN111418010B (en) | Multi-microphone noise reduction method and device and terminal equipment | |
| US9343056B1 (en) | Wind noise detection and suppression | |
| US8965757B2 (en) | System and method for multi-channel noise suppression based on closed-form solutions and estimation of time-varying complex statistics | |
| CN106340292B (en) | A kind of sound enhancement method based on continuing noise estimation | |
| CA2607981C (en) | Multi-sensory speech enhancement using a clean speech prior | |
| AU2005202858A1 (en) | Method and apparatus for multi-sensory speech enhancement | |
| TWI581254B (en) | Environmental noise elimination system and application method thereof | |
| CN110556125B (en) | Feature extraction method and device based on voice signal and computer storage medium | |
| AU2009203194A1 (en) | Noise spectrum tracking in noisy acoustical signals | |
| CN101233561B (en) | Enhancement of speech intelligibility in a mobile communication device by controlling the operation of a vibrator of a vibrator in dependance of the background noise | |
| CN110931027A (en) | Audio processing method and device, electronic equipment and computer readable storage medium | |
| CN110875054B (en) | Far-field noise suppression method, device and system | |
| CN107452398B (en) | Echo acquisition method, electronic device and computer readable storage medium | |
| CN103824563A (en) | Hearing aid denoising device and method based on module multiplexing | |
| JP3786038B2 (en) | Input signal processing method and input signal processing apparatus | |
| CN112165558B (en) | Method and device for detecting double-talk state, storage medium and terminal equipment | |
| CN112802490B (en) | Beam forming method and device based on microphone array | |
| CN113744752A (en) | Voice processing method and device | |
| CN113035222A (en) | Voice noise reduction method and device, filter determination method and voice interaction equipment | |
| KR20110024969A (en) | Apparatus for filtering noise by using statistical model in voice signal and method thereof | |
| CN114495964B (en) | Frequency domain filtering voice noise reduction method based on pearson correlation coefficient | |
| CN113611321A (en) | Voice enhancement method and system | |
| CN112669877B (en) | Noise detection and suppression method and device, terminal equipment, system and chip |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |