CN107564539A - Towards the acoustic echo removing method and device of microphone array - Google Patents
Towards the acoustic echo removing method and device of microphone array Download PDFInfo
- Publication number
- CN107564539A CN107564539A CN201710757604.4A CN201710757604A CN107564539A CN 107564539 A CN107564539 A CN 107564539A CN 201710757604 A CN201710757604 A CN 201710757604A CN 107564539 A CN107564539 A CN 107564539A
- Authority
- CN
- China
- Prior art keywords
- echo
- signal
- microphone array
- module
- microphone
- 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
Landscapes
- Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
- Circuit For Audible Band Transducer (AREA)
Abstract
The present invention relates to a kind of acoustic echo removing method towards microphone array, microphone array gathers the data of near end signal and echo signal in real time, choose the channel signal of one of microphone, according to adaptive-filtering thought and independent composition analysis algorithm structure initialization model, echo cancellor model is established according to initialization model echo cancellor is carried out to microphone single channel signal, model is eliminated according to initial echo and single pass filter coefficient goes the echo cancellor estimation model for building other microphones in microphone array to carry out echo cancellor, the present invention can carry out acoustic echo elimination to the microphone array of multichannel, effective data processing is provided and improves phonetic recognization rate；Device provided by the invention passes through the key position that is arranged in acoustic sensor on one loop sensor array, the multichannel recording signal and echo signal of measuring system carry out echo cancellor, the echo cancellation signal of microphone array is obtained, is finally reached the purpose for improving phonetic recognization rate.
Description
Technical field
The present invention relates to a kind of digital processing field, more particularly to a kind of acoustic echo towards microphone array to disappear
Except method and device.
Background technology
With the arriving in artificial intelligence epoch, interface of the voice technology as man-machine interaction, increasingly closed by people
Note.Traditional near field voice interaction technique can not meet the needs of people, people be more desirable to development can it is farther away from
From with Voice command smart machine in more complicated scene.Thus, microphone array technology turns into the core of far field interactive voice.
So-called microphone array, that is, refer to the acoustic sensor (microphone) of certain amount by the multiple of necessarily regularly arranged
Microphone system, and applied to the processing of voice signal.Conventional interactive voice application scenarios, mainly using single microphone, but
Single microphone systems are needed sound source is relatively near, could be obtained under low noise, the recent developments without reverberation and meets the letter of speech recognition requirement
Number, therefore there is certain limitation.For the application scenarios of far field speech recognition, largely made an uproar due to being included in true environment
Sound, reverberation and echo, the signal quality for causing microphone system to pick up is poor, has a strong impact on the discrimination of voice.And due to
The problem of more sound sources and ambient noise, single microphone systems are difficult to the separation and positioning of each sound source, it is therefore desirable to more Mikes
Wind system carrys out the front end signal processing of assistant voice identification.
For current complicated application scenarios, having been developed based on microphone array a series of can effectively improve speech recognition
The key technology of rate, mainly includes：Speech enhan-cement, auditory localization, reverberation elimination, echo cancellor, noise suppressed.Wherein, voice
Enhancing technology mainly uses Beamforming Method, and it mainly strengthens the sound of specific direction, suppresses the sound interference beyond main lobe；
Auditory localization, refers mainly to calculate the angle and distance of target speaker using microphone array, and then realizes target speaker
Tracking and sound orientation pickup；Reverberation eliminates, and is mainly disappeared using the signal pre-treatment means such as liftering or linear prediction
Except the nonsynchronous voice for influenceing speech recognition is overlapped mutually phenomenon；Echo cancellor, mainly utilize the hands such as Adaptive Signal Processing
Phenomena such as section is to eliminate the stress in interference speech recognition；Noise suppressed, mainly eliminate language using noise-reduction methods such as spectrum-subtractions
Larger ambient noise be present under sound identification application scenarios.
Based on above-mentioned engineering background, one of key technology of the echo cancellor as microphone array, to microphone array
Multichannel acoustic echo cancellation algorithm under row can provide important technical support for far field interactive voice.
The content of the invention
In order to solve the above technical problems, it is an object of the invention to eliminate difficult ask for far field speech recognition acoustic echo
A kind of topic, there is provided acoustics towards microphone array being combined based on independent composition analysis algorithm and adaptive-filtering thought
Echo cancel method and device.
Acoustic echo removing method provided by the invention towards microphone array, microphone array gather near-end letter in real time
Number and echo signal data；
The microphone channel signal of one of microphone output is chosen, according to adaptive-filtering thought and independent element point
Algorithm structure initialization model is analysed, echo cancellor model is established according to the initialization model microphone channel signal is entered
Row echo cancellor；
Model is eliminated according to initial echo and single pass filter coefficient goes to build other Mikes in microphone array
The echo cancellor estimation model of wind carries out echo cancellor.
Further, the microphone channel signal includes the near end signal and echo signal that the microphone is collected.
Further, the initialization model includes the echo signal and remote signaling in the microphone channel signal
Functional relation and the functional relation of the echo signal to estimating in other described microphone channel signals and remote signaling difference
For
With
WhereinFor impulse response of the remote signaling through spatial,For echo signal,For the echo of estimation
Signal,For remote signaling,For the adaptive-filtering coefficient of estimation.
Further, the echo signal of the estimation is with controlling the relation of the convergent error signal of adaptive-filtering
WhereinBelieve for the microphone channel
Number.
Further, the echo cancellor model is
WhereinFor the near end signal.
Further, the echo cancellor estimation model is
WhereinIt is non-zero scalar.
More specifically, it is describedSolved by Fast ICA.
A kind of acoustic echo cancellation element towards microphone array provided by the invention, including central processing module, adopt
Collect module, memory module, power module, Upper machine communication module, reseting module and control key module, the central processing module is used
In controlling the acquisition module and handling the signal of the acquisition module collection, the memory module is used to store the centre
The signal that reason device inputs the signal of the memory module and the memory module exports to the central processing module, it is described on
Position machine communication module is used to handle the signal handled through the central processing module and controls the central processing module, described multiple
Position module is used for the reset for controlling the central processing module, and the control key module, which is used to input to the central processing module, joins
Number, via such scheme, the present invention at least has advantages below：A kind of acoustics towards microphone array provided by the invention returns
Sound removing method and device can realize the echo cancellor under multichannel, improve phonetic recognization rate.
Further, the CPU module is the OMAP dual core processors based on ARM cores and DSP core.
Further, the acquisition module is microphone array data acquisition module.
Further, the Upper machine communication module also includes being used for the cloud server for carrying out speech recognition.
Acoustic echo removing method and device provided by the invention towards microphone array, can be to multi-microphone array system
System carries out effective multi-channel echo elimination, provides effective pre-processing algorithm technical support for rear end speech recognition, and apply
In on the devices such as intelligent robot, intelligent sound box, intelligent meeting or system.
Described above is only the general introduction of technical solution of the present invention, in order to better understand the technological means of the present invention,
And can be practiced according to the content of specification, below with presently preferred embodiments of the present invention and coordinate accompanying drawing describe in detail as after.
Brief description of the drawings
Fig. 1 is the algorithm principle block diagram of the acoustic echo removing method provided by the invention towards microphone array；
Fig. 2 is the structured flowchart of the acoustic echo cancellation element provided by the invention towards microphone array.
Embodiment
With reference to the accompanying drawings and examples, the embodiment of the present invention is described in further detail.Implement below
Example is used to illustrate the present invention, but is not limited to the scope of the present invention.
Referring to Fig. 1, Fig. 1 is a kind of the one preferable of acoustic echo removing method towards microphone array provided by the invention
Embodiment.In the present embodiment：
First, microphone array gathers the data of echo signal near end signal and echo passage in recording channel in real time,
Then the microphone channel signal of one of microphone output is chosen, according to the data in the microphone channel signal, foundation
The initialization model of adaptive-filtering thought and independent composition analysis algorithm structure ICA, is established back according to the ICA initialization models
Sound eliminates model and carries out echo cancellor to microphone channel signal, and model and single pass wave filter are eliminated according to initial echo
Coefficient goes the echo cancellor estimation model for building other microphones in microphone array to carry out echo cancellor.
Preferably, comprise the following steps that：
S1, set measured remote signaling asIf the signal D of microphone array collectionkFor
Wherein, the sonic transducer number that m expressions are arranged on microphone structure, k expression time domain sampling point numbers, 1≤j≤
m；1≤i≤k；
S2, set near-end voice signals asEcho signal isTherefore j-th of microphone channel signal, near end signal,
Echo signal meets following relation：
S3, set impulse response of the remote signaling through spatial asTherefore echo signalWith remote signaling's
Relation is：
S4, set the echo signal of estimation asThe adaptive-filtering coefficient of estimation isTherefore the echo signal that band is estimated
With remote signaling relation and the control convergent error signal of adaptive-filteringRelation is：
S5, based on ICA thoughts, structure acoustic echo eliminates ICA models, and model equation is represented by：
Wherein,For required original signal,For observation vector,For hybrid matrix；
S6, the equation for separating primary signal can be established according to S05, can represented as follows：
Wherein,It is non-zero scalar, obtains
S7, in order to as far as possible separate target voice, solved using Fast ICA methodsComprise the following steps that：
(a) to observing data matrixCarry out equalization；
(b) setAlbefaction observation data matrix x (Z=vx), wherein v=D-1/2ETTo be white
Change matrix, D, E are respectively to observe characteristic value and characteristic vector corresponding to data matrix x covariance matrix；
(c) ICA component numbers to be separated, juxtaposition counter l=1 are selected；
(d) initial random weight coefficient is set
(e) renewal solves
(f) decorrelation：
(g) weight coefficient is normalized：
If (h)Do not restrainReturn to step (e)；
(i) iteration l=l+1, if l≤m, return to step (d)；
S8, for+1 passage of jth, the weight coefficient tried to achieve using S07The approximate echo cancellor for seeking other passages is removed,
Its equation is as follows：
The estimation calculation formula for obtaining the near end signal in other microphone channel signals is as follows：
So, this method using independent component analysis (Independent Component Analysis, ICA) algorithm and
Adaptive-filtering thought structure echo cancellor estimation model, separation target voice is from remote signaling and microphone pickup signals
Near end signal, the target voice of separation is finally transferred to identification end, and then echo cancellor and the purpose of speech recognition, compared with
In tradition multi-channel echo elimination under need that to each passage echo cancellation algorithm will be carried out, effectively save amount of calculation
And algorithm complex.
Referring to Fig. 2, Fig. 2 is a kind of the one preferable of acoustic echo cancellation element towards microphone array provided by the invention
Embodiment, for realizing the above-mentioned acoustic echo removing method towards microphone array.In the present embodiment：
Acoustic echo cancellation element towards microphone array includes central processing module, acquisition module, memory module, electricity
Source module, Upper machine communication module, reseting module and keying module, central processing module are used to control acquisition module and processing to adopt
Collect the signal of module collection, memory module is used for the signal for storing central processing unit input memory module and memory module in
The signal of processing module output is entreated, Upper machine communication module is used to handle the signal handled through central processing module and controls center
Processing module, reseting module are used for the reset for controlling central processing module, and keying module, which is used to input to central processing module, joins
Number.
Preferably, in order that central processing module small power consumption, data-handling capacity are strong, central processing module is double to possess
The OMAP dual core processors of core ARM cores and DSP core structure.
Preferably, acquisition module is the input with signal, the conditioning of signal, the function of A/D Data collecting conversions
Microphone array data acquisition module.
Preferably, Upper machine communication module also includes being used for the cloud server for carrying out speech recognition, so, host computer
Data after echo cancellor can be carried out speech recognition by communication module through the identification end in network transmission to cloud server.
Job step towards the acoustic echo cancellation element of microphone array is as follows：
First, via Upper machine communication module or keying module to the working condition of device, passage control, collection class signal
Type, sample frequency etc. carry out parameter setting, parameter instruction are transmitted into central processing module, driving acquisition module enters to speech data
Row collection；Then, what the echo channel signal and recording channel that the DSP module of OMAP processors gathers to microphone array gathered
Mixed signal carries out echo cancellor；Finally, by high in the clouds of the data after echo cancellor through network transmission to Upper machine communication module
The speech recognition system of server.
Acoustic echo cancellation element provided by the invention towards microphone array is that multiple sonic transducers are arranged in into ring
In the key point of shape microphone array array structure, echo cancellor is carried out by the voice signal obtained to measurement, system entered in real time
Row echo cancellor, and be applied in the application of intelligent sound box or intelligent robot.Described multi-channel echo cancellation element
Using OMAP embeded processors as core, the units such as data acquisition, data storage, control and data processing are integrated with, the device
Design take full advantage of OMAP double-core ARM cores and DSP core structure, have that ARM cores are low in energy consumption, processing speed is fast, task is adjusted
The functions such as flexible mode are spent, and the function of the powerful digital processing analysis of DSP core, by the two efficient combination to realize that voice is believed
The collection of number real-time online, processing, transmission and analysis.Carried out data transmission simultaneously using Ethernet, realize the quick, efficient of data
Transmission, avoids the loss of signal in the transmission, accomplishes resource-sharing, the shortcomings of overcoming traditional data collection offline, delay.
It the above is only the preferred embodiment of the present invention, be not intended to limit the invention, it is noted that led for this technology
For the those of ordinary skill in domain, without departing from the technical principles of the invention, some improvement and modification can also be made,
These improvement and modification also should be regarded as protection scope of the present invention.
Claims (10)
- A kind of 1. acoustic echo removing method towards microphone array, it is characterised in that：Microphone array gathers near end signal and echo signal data in real time；The microphone channel signal of one of microphone collection is chosen, is calculated according to adaptive-filtering thought and independent component analysis Method builds initialization model, establishes echo cancellor model according to the initialization model and the microphone channel signal is returned Sound eliminates；Model is eliminated according to initial echo and single pass filter coefficient goes to build other microphones in microphone array Echo cancellor estimation model carry out echo cancellor.
- 2. the acoustic echo removing method according to claim 1 towards microphone array, it is characterised in that：The Mike Wind passage includes the near end signal and echo signal that the microphone is collected.
- 3. the acoustic echo removing method according to claim 1 towards microphone array, it is characterised in that：It is described initial Change model including the echo signal in the microphone channel signal and the functional relation of remote signaling and to other described wheats The echo signal and the functional relation of remote signaling estimated in gram wind channel signal be respectivelyWithWhereinFor impulse response of the remote signaling through spatial,For echo signal,For the echo signal of estimation,For remote signaling,For the adaptive-filtering coefficient of estimation.
- 4. the acoustic echo removing method according to claim 3 towards microphone array, it is characterised in that：The estimation Echo signal be with controlling the relation of the convergent error signal of adaptive-filteringWhereinFor the microphone channel signal.
- 5. the acoustic echo removing method according to claim 1 towards microphone array, it is characterised in that：The echo Eliminating model isWhereinFor the near end signal.
- 6. the acoustic echo removing method according to claim 5 towards microphone array, it is characterised in that：The echo Eliminating estimation model isWhereinIt is non-zero scalar.
- 7. the acoustic echo removing method according to claim 6 towards microphone array, it is characterised in that：It is describedIt is logical Fast ICA are crossed to be solved.
- A kind of 8. acoustic echo cancellation element towards microphone array, it is characterised in that：Including central processing module, collection mould Block, memory module, power module, Upper machine communication module, reseting module and keying module, the central processing module are used to control Make the acquisition module and handle the signal of the acquisition module collection, the memory module is used to store the central processing unit Input the signal of the memory module and signal that the memory module exports to the central processing module, the host computer Communication module is used to handle the signal handled through the central processing module and controls the central processing module, the reset mould Block is used for the reset for controlling the central processing module, and the keying module is used for the central processing module input parameter.
- 9. the acoustic echo cancellation element according to claim 8 towards microphone array, it is characterised in that：The center Processor module is the OMAP dual core processors based on ARM cores and DSP core, and the acquisition module is microphone array data acquisition Module.
- 10. the acoustic echo cancellation element according to claim 8 towards microphone array, it is characterised in that：On described Position machine communication module also includes being used for the cloud server for carrying out speech recognition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710757604.4A CN107564539B (en) | 2017-08-29 | 2017-08-29 | Acoustic echo cancellation method and device facing microphone array |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710757604.4A CN107564539B (en) | 2017-08-29 | 2017-08-29 | Acoustic echo cancellation method and device facing microphone array |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107564539A true CN107564539A (en) | 2018-01-09 |
| CN107564539B CN107564539B (en) | 2021-12-28 |
Family
ID=60977477
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710757604.4A Active CN107564539B (en) | 2017-08-29 | 2017-08-29 | Acoustic echo cancellation method and device facing microphone array |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107564539B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108305631A (en) * | 2018-04-04 | 2018-07-20 | 西安合谱声学科技有限公司 | A kind of Acoustic treatment equipment based on multinuclear modularization framework |
| CN110265054A (en) * | 2019-06-14 | 2019-09-20 | 深圳市腾讯网域计算机网络有限公司 | Audio signal processing method, device, computer readable storage medium and computer equipment |
| CN110782911A (en) * | 2018-07-30 | 2020-02-11 | 阿里巴巴集团控股有限公司 | Audio signal processing method, apparatus, device and storage medium |
| CN111031448A (en) * | 2019-11-12 | 2020-04-17 | 西安讯飞超脑信息科技有限公司 | Echo cancellation method, echo cancellation device, electronic equipment and storage medium |
| CN111261179A (en) * | 2018-11-30 | 2020-06-09 | 阿里巴巴集团控股有限公司 | Echo cancellation method and device and intelligent equipment |
| CN111312269A (en) * | 2019-12-13 | 2020-06-19 | 辽宁工业大学 | Rapid echo cancellation method in intelligent loudspeaker box |
| CN112216295A (en) * | 2019-06-25 | 2021-01-12 | 大众问问(北京)信息科技有限公司 | Sound source positioning method, device and equipment |
| CN112236820A (en) * | 2018-06-25 | 2021-01-15 | 赛普拉斯半导体公司 | Beamformer and Acoustic Echo Canceller (AEC) system |
| CN115696140A (en) * | 2022-12-05 | 2023-02-03 | 长沙东玛克信息科技有限公司 | Classroom audio multi-channel echo cancellation method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1750573A (en) * | 2004-09-03 | 2006-03-22 | 哈曼贝克自动***股份有限公司 | Speech signal processing with combined noise reduction and echo compensation |
| CN101820302A (en) * | 2009-02-27 | 2010-09-01 | 比亚迪股份有限公司 | Device and method for canceling echo |
| US20110022361A1 (en) * | 2009-07-22 | 2011-01-27 | Toshiyuki Sekiya | Sound processing device, sound processing method, and program |
| CN103426436A (en) * | 2012-05-04 | 2013-12-04 | 索尼电脑娱乐公司 | Source separation by independent component analysis in conjuction with optimization of acoustic echo cancellation |
| CN104883462A (en) * | 2014-02-28 | 2015-09-02 | 中国科学院声学研究所 | Adaptive filter and filtering method used for eliminating acoustics echo |
| CN106875938A (en) * | 2017-03-10 | 2017-06-20 | 南京信息工程大学 | A kind of improved nonlinear adaptive sound end detecting method |
-
2017
- 2017-08-29 CN CN201710757604.4A patent/CN107564539B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1750573A (en) * | 2004-09-03 | 2006-03-22 | 哈曼贝克自动***股份有限公司 | Speech signal processing with combined noise reduction and echo compensation |
| CN101820302A (en) * | 2009-02-27 | 2010-09-01 | 比亚迪股份有限公司 | Device and method for canceling echo |
| US20110022361A1 (en) * | 2009-07-22 | 2011-01-27 | Toshiyuki Sekiya | Sound processing device, sound processing method, and program |
| CN103426436A (en) * | 2012-05-04 | 2013-12-04 | 索尼电脑娱乐公司 | Source separation by independent component analysis in conjuction with optimization of acoustic echo cancellation |
| CN104883462A (en) * | 2014-02-28 | 2015-09-02 | 中国科学院声学研究所 | Adaptive filter and filtering method used for eliminating acoustics echo |
| CN106875938A (en) * | 2017-03-10 | 2017-06-20 | 南京信息工程大学 | A kind of improved nonlinear adaptive sound end detecting method |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108305631B (en) * | 2018-04-04 | 2023-09-05 | 西安合谱声学科技有限公司 | Acoustic processing equipment based on multi-core modularized framework |
| CN108305631A (en) * | 2018-04-04 | 2018-07-20 | 西安合谱声学科技有限公司 | A kind of Acoustic treatment equipment based on multinuclear modularization framework |
| CN112236820A (en) * | 2018-06-25 | 2021-01-15 | 赛普拉斯半导体公司 | Beamformer and Acoustic Echo Canceller (AEC) system |
| CN112236820B (en) * | 2018-06-25 | 2024-04-26 | 赛普拉斯半导体公司 | Beamformer and Acoustic Echo Canceller (AEC) system |
| CN110782911A (en) * | 2018-07-30 | 2020-02-11 | 阿里巴巴集团控股有限公司 | Audio signal processing method, apparatus, device and storage medium |
| CN111261179A (en) * | 2018-11-30 | 2020-06-09 | 阿里巴巴集团控股有限公司 | Echo cancellation method and device and intelligent equipment |
| CN110265054B (en) * | 2019-06-14 | 2024-01-30 | 深圳市腾讯网域计算机网络有限公司 | Speech signal processing method, device, computer readable storage medium and computer equipment |
| CN110265054A (en) * | 2019-06-14 | 2019-09-20 | 深圳市腾讯网域计算机网络有限公司 | Audio signal processing method, device, computer readable storage medium and computer equipment |
| CN112216295A (en) * | 2019-06-25 | 2021-01-12 | 大众问问(北京)信息科技有限公司 | Sound source positioning method, device and equipment |
| CN112216295B (en) * | 2019-06-25 | 2024-04-26 | 大众问问(北京)信息科技有限公司 | Sound source positioning method, device and equipment |
| CN111031448B (en) * | 2019-11-12 | 2021-09-17 | 西安讯飞超脑信息科技有限公司 | Echo cancellation method, echo cancellation device, electronic equipment and storage medium |
| CN111031448A (en) * | 2019-11-12 | 2020-04-17 | 西安讯飞超脑信息科技有限公司 | Echo cancellation method, echo cancellation device, electronic equipment and storage medium |
| CN111312269A (en) * | 2019-12-13 | 2020-06-19 | 辽宁工业大学 | Rapid echo cancellation method in intelligent loudspeaker box |
| CN111312269B (en) * | 2019-12-13 | 2023-01-24 | 天津职业技术师范大学(中国职业培训指导教师进修中心) | Rapid echo cancellation method in intelligent loudspeaker box |
| CN115696140A (en) * | 2022-12-05 | 2023-02-03 | 长沙东玛克信息科技有限公司 | Classroom audio multi-channel echo cancellation method |
| CN115696140B (en) * | 2022-12-05 | 2023-05-26 | 长沙东玛克信息科技有限公司 | Classroom audio multichannel echo cancellation method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107564539B (en) | 2021-12-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107564539A (en) | Towards the acoustic echo removing method and device of microphone array | |
| CN107333120B (en) | Integrated sensor based on microphone array and stereoscopic vision | |
| CN109800700B (en) | Underwater acoustic signal target classification and identification method based on deep learning | |
| CN102831898B (en) | Microphone array voice enhancement device with sound source direction tracking function and method thereof | |
| CN108109617B (en) | Remote pickup method | |
| CN111161751A (en) | Distributed microphone pickup system and method under complex scene | |
| CN111044973B (en) | MVDR target sound source directional pickup method for microphone matrix | |
| CN109074816A (en) | Far field automatic speech recognition pretreatment | |
| CN206349145U (en) | Audio signal processing apparatus | |
| CN107742522A (en) | Target voice acquisition methods and device based on microphone array | |
| CN106782584A (en) | Audio signal processing apparatus, method and electronic equipment | |
| CN104157293A (en) | Signal processing method for enhancing target voice signal pickup in sound environment | |
| JP2019168730A (en) | Estimating room impulse response for acoustic echo cancellation | |
| CN108389586A (en) | A kind of long-range audio collecting device, monitoring device and long-range collection sound method | |
| CN110444220B (en) | Multi-mode remote voice perception method and device | |
| WO2022012206A1 (en) | Audio signal processing method, device, equipment, and storage medium | |
| CN109979476A (en) | A kind of method and device of speech dereverbcration | |
| CN111312269B (en) | Rapid echo cancellation method in intelligent loudspeaker box | |
| US20220201421A1 (en) | Spatial audio array processing system and method | |
| CN110120217A (en) | A kind of audio data processing method and device | |
| CN106992010A (en) | Without the microphone array speech enhancement device under the conditions of direct sound wave | |
| CN108337605A (en) | The hidden method for acoustic formed based on Difference Beam | |
| CN110515034B (en) | Acoustic signal azimuth angle measurement system and method | |
| Li et al. | The PCG-AIID system for L3DAS22 challenge: MIMO and MISO convolutional recurrent network for multi channel speech enhancement and speech recognition | |
| CN112180318B (en) | Sound source direction of arrival estimation model training and sound source direction of arrival estimation method |
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 |