67results about How to "Improve echo cancellation effect" patented technology

The invention relates to an echo cancellation method, an echo cancellation device, a conference tablet computer and a computer storage medium. The method comprises steps: reference signals inputted to a loudspeaker for playback are acquired, and acquisition signals of a microphone are acquired; coefficients of an adaptive filter are acquired, and the adaptive filter coefficients are used for estimating echo signals corresponding to the reference signals; the echo signals are canceled from the acquisition signals, and residual signals are acquired and outputted; and according to the relativity between the residual signals and the reference signals, the coefficients of the adaptive filter are updated. The embodiment creatively adjusts the coefficients of the adaptive filter through the relativity between the residual signals and the reference signals, echo cancellation can be carried out to an extremely large degree, updating on the coefficients of the adaptive filter only needs to be added on the basis of the traditional algorithm, and the newly-added cost is low.

The invention discloses an echo elimination method based on deep learning, an echo elimination device based on deep learning, electronic equipment and storage medium, and belongs to the technical field of a computer. The method comprises the following steps of extracting acoustic features from received microphone signals, wherein the microphone signals comprise near-end signals and remote-end signals; performing iterative operation on the acoustic features in a pretrained recursive neural network model with long short-term memory; calculating a ratio membrane of the acoustic features; using the ratio membrane for shielding the acoustic features; and synthesizing the shielded acoustic features with the phase of the microphone signals to obtain near end signals subjected to echo elimination.The echo elimination method based on deep learning and the echo elimination device based on deep learning can realize the echo elimination under the conditions of the background noise, double talking, nonlinear distortion and the like; the echo elimination effect is greatly improved; the application scenes are greatly increased; the use of a post-filter is not needed; the electronic equipment iseffectively simplified; and the electronic equipment cost is reduced.

The related echo eliminator comprises: a linear predicted decoder to decode data into remote speech signal and extract the short activating signal and linear prediction coding parameter, a decorrelation filter to receive the predicted coding parameter and error signal for decorrelation and generate decorrelated error signal, an adaptive filter to receive all of the remote speech and short activating signal and decorrelated error signal and generate the estimated echo, and a subtracter to generate the error signal. This invention combines with LPC device, reduces algorithm complexity, and prompts algorithm convergence.

The invention discloses a proportional adaptive filter coefficient vector updating method using coefficient reusing. According to the method, a current filter coefficient vector and each Euclidean norm of last R filter coefficient vector differences to sum-of-squares minimization, under the constraint condition that posteriori error is zero; a diagonal distribution matrix G(n) is introduced, and diagonal elements of the matrix correspond to proportional step size of each coefficient. When the method is applied to estimating a coefficient unknown system, convergence rate is higher and stability maladjustment is lower.

The invention discloses an echo cancellation method and device based on time delay estimation. The method comprises the steps that microphone signals and reference signals are acquired and preprocessed; frequency-point signals with nonlinear conditions in the preprocessed microphone signals and reference signals in a current scene are determined; according to frequency-point signals without nonlinear conditions in the microphone signals and the reference signals, a current time delay estimation value is obtained through calculation; based on the current time delay estimation value, the reference signals are moved; according to the preprocessed microphone signals and the moved reference signals, a self-adaptive filter is updated, and echo cancellation is achieved. By means of the method anddevice, the accuracy of time delay estimation can be improved, and the echo cancellation effect is improved.

The invention provides a method and a device for echo cancellation, wherein a signal strength value of a distal end signal is used as one of parameters for adjusting a simulation echo signal transfer function; nonlinear factors are led in when the signal strength of the distal end signal exceeds the strength threshold of an abnormal signal, so that the convergence speed of the echo simulation algorithm is increased, and the problem that non-convergence is possible to be caused by too high signal strength of the distal end signal is avoided; in addition, the adaptive capacity of the method forecho cancellation on the environment is improved, so that the effect of echo cancellation can be improved.

The invention discloses a speech signal time delay estimation method and system used for echo cancellation. The method comprises the following steps of respectively obtaining a far-end signal and a near-end signal of a voice signal received by a microphone in a call and generating a corresponding far-end frequency domain signal and a corresponding near-end frequency domain signal; adopting a critical band based on the auditory masking effect to respectively carry out sub-band decomposition in the frequency band on the far-end frequency domain signal and the near-end frequency domain signal and obtaining a power spectrum of each critical frequency point of the far-end frequency domain signal and a power spectrum of each critical frequency point of the near-end frequency domain signal; respectively extracting local binary features of the power spectrum of each critical frequency point of the far-end frequency domain signal and the power spectrum of each critical frequency point of the near-end frequency domain signal, respectively matching and generating and outputting a time delay estimation result. According to the method and the system, the correlation between the far-end signal and the near-end signal is counted based on the auditory masking effect, the computation is simple, and the accurate time delay estimation can be obtained to bring great convenience for echo cancellation.

The invention discloses a sound pickup adjustment method, which comprises the following steps: obtaining the orientation information of a sound source; determining an angle of rotation of the loudspeaker according to the azimuth information. The gain corresponding to each microphone is determined according to the determined angle and the position of the plurality of microphones, and the pickup parameter of the corresponding microphone is updated according to the determined gain. The invention also discloses a control terminal and a computer-readable storage medium. A sound pickup adjustment method of the invention realizes that the sound pickup signals recorded by different microphones tend to be close to each other or be adjusted to be consistent. Thus, an ideal speech input is provided for the intelligent speech recognition algorithm of the intelligent sound box, the performance of the speech recognition of the intelligent sound box is guaranteed from the source, and the positioningaccuracy of the sound source of the intelligent sound box and the effect of the echo cancellation are improved.

An acoustic echo removing apparatus detects space information representing a location of a sound source using a plurality of microphone input signals that are received through a plurality of microphones, and generates an acoustic echo estimation signal from a far-end talker voice signal using an adaptive filter coefficient. The acoustic echo removing apparatus detects a double talk segment using the space information, and determines update of the adaptive filter coefficient according to whether the double talk segment exists.

The embodiment of the invention provides an echo cancellation method and device and an electronic device. The method comprises the steps that currently acquired audio data acquired by an audio acquisition module are acquired; the current moment and propagation time delay are acquired; a search range is determined according to the current moment and the propagation time delay; backup audio data corresponding to an echo signal of the currently acquired audio data are searched for in the backup audio data according to the search range to serve as an echo cancellation reference frame, wherein thebackup audio data are stored after the electronic device copies audio data output to an audio playing module; and echo signal cancellation is performed on the currently acquired audio data according to the echo cancellation reference frame. The search range for searching for the reference frame is determined according to the propagation time delay obtained by the electronic device through testingin advance, and the search range is more accurate, so that the echo cancellation effect can be improved.

The invention discloses an echo elimination device of a law enforcement recorder. The echo elimination device comprises a far-end user, a far-end user voice input device, a far-end user voice output device, a first signal converter, a radio transmission device, a voice processing device, a second signal converter, a local user voice output device, a local user voice input device and a local user;the far-end user is connected through the far-end user voice input device, the first signal converter and the radio transmission device; the radio transmission device is connected with the local userthrough the voice processing device, the second signal converter and the far-end user voice output device; and the output end of the local user is connected with the input end of the second signal converter through the local user voice input device. The echo elimination device of the law enforcement recorder of the present invention is not interfered by the output data of the far-end user in the local user voice input device, ensures the effect of echo cancellation during a certain period, and improves the efficiency and effectiveness of echo cancellation on the whole.

The invention discloses an echo cancellation method combining voiceprint recognition and deep learning, and the method comprises the steps: 1) extracting acoustic features from a voice segment of a target speaker, and obtaining an embedded vector, i.e., a voiceprint vector, representing the information of the speaker through a pre-trained voiceprint encoder comprising a recurrent neural network model; and 2) extracting the acoustic features from the received signals, wherein the signals comprise near-end signals and far-end signals, estimating appearing echo signals, namely estimated echo signals through an AEC algorithm, and meanwhile obtaining near-end signals, namely linear filtering signals, containing residual echoes and eliminated in a linear mode. According to the echo cancellationmethod combining voiceprint recognition and deep learning provided by the invention, echo cancellation can be realized under the conditions of double speaking, non-occurrence distortion and the like,and the echo cancellation effect is greatly improved even under the condition that a far-end signal contains human voice.

The invention relates to an echo elimination method, an apparatus and a system, and a computer readable storage medium. The method comprises the following steps of acquiring and estimating characteristic parameters of a local signal, wherein the characteristic parameters include a current frame reference signal emitted by a sound production device, a signal received by a current frame microphone and a previous frame adaptive filter coefficient; converting the characteristic parameters into frequency domain signals respectively; according to the converted frequency domain signals, estimating the local signal; calculating a variance of the local signal; and according to the variance of the local signal, updating a step length factor and a regularization factor of an adaptive filter. Throughcarrying out FFT(fast Fourier transform) on the acquired characteristic parameters used for estimating the local signal, time domain signals are converted into the frequency domain signals. Sparsity and distinguishability of the frequency domain signals are used so that the local signal estimated through the frequency domain signals and the calculated variance of the local signal are accurate, which is good for enhancing an echo elimination effect.

The invention relates to the technical field of voice communication, in particular to an audio signal processing method and a device thereof. The audio signal processing method comprises the following steps: acquiring a first audio signal picked up by a first microphone and a second audio signal picked up by a second microphone; wherein each of the first audio signal and the second audio signal comprises an echo signal generated by playing a reference signal by a loudspeaker; respectively carrying out linear filtering processing on the first audio signal and the second audio signal to obtain a first residual signal and a second residual signal; and performing nonlinear filtering processing on the second residual signal according to the first residual signal to obtain a target voice signal. According to the method, linear and nonlinear echo signals can be eliminated, the echo elimination effect is improved, near-end voice signals are approximately lossless, and the voice communication quality is improved.

The embodiment provides a display device, a multichannel echo cancellation circuit and a multichannel echo cancellation method, the display device comprises a loudspeaker, the loudspeaker comprises a main sound channel and an auxiliary sound channel, the main sound channel plays sound according to a first audio after frequency division, and the auxiliary sound channel plays sound according to a second audio after frequency division. The first audio is obtained by executing first processing on an original audio obtained from the playing service processor, the second audio is obtained by executing second processing on the first audio, and the first processing and the second processing are different audio processing modes. And the controller performs echo cancellation on the first echo and the second echo according to the recovery reference signal corresponding to the second audio. The second audio comprises the original audio, the related information of the first processing and the related information of the second processing, so that the purpose of completely eliminating the echo of multiple sound channels can be achieved, the echo elimination effect is improved, the voice interaction effect is finally improved, and the user experience is improved.

The invention provides an echo cancellation method and electronic equipment, wherein the method comprises the steps: processing a to-be-processed voice signal at a current moment by an AEC algorithm,and obtaining a first voice signal; inputting the first voice signal into a pre-trained neural network model to obtain a ratio film of the first voice signal; and processing the first voice signal byusing the ratio film, and adjusting the parameters of the AEC algorithm based on the ratio film so as to process the to-be-processed voice signal at the next moment based on the AEC algorithm after parameter adjustment. According to the embodiment of the invention, the echo cancellation effect can be improved.

The invention discloses a recurrent neural network model training method, an echo cancellation method, terminal equipment and a computer readable storage medium. The method comprises the steps: obtaining a sample signal which comprises a microphone signal and a reference signal corresponding to the microphone signal; carrying out feature extraction on the sample signal to obtain a first acoustic feature and a second acoustic feature, wherein the first acoustic feature is the acoustic feature of the microphone signal, and the second acoustic feature is the acoustic feature of the reference signal; inputting the first acoustic feature and the second acoustic feature into an initial recurrent neural network model, and outputting a predicted value of a phase sensitive mask, wherein the phase sensitive mask is a frequency spectrum proportion of an echo cancellation signal in the microphone signal, and a frequency spectrum comprises an amplitude spectrum and a phase spectrum; and calculating loss of the phase sensitive mask to adjust parameters of the initial recurrent neural network model to obtain the recurrent neural network model. In this way, the echo cancellation capability can be improved.

The invention discloses an echo cancellation method, and belongs to the technical field of sound processing. The method comprises the following steps of acquiring a near-end sound signal and a far-endsound signal, wherein each of the near-end sound signal and the far-end sound signal is composed of multiple pieces of sampling point data, and the sampling point data are f-bit data; inputting the acquired near-end sound signal and far-end sound signal into an echo cancellation algorithm model; and by taking fXh-bit data as units, performing echo cancellation processing on the input signals through the echo cancellation algorithm model, and outputting the sound signals subjected to the echo cancellation, wherein h is an integer greater than or equal to 2. According to the scheme, for the signals organized by the f-bit sampling point data, parallel processing is performed by taking the h pieces of sampling point data as a group, so that the duration required for processing the sound signals is shortened, the echo cancellation efficiency can be effectively improved, and the effect of performing the echo cancellation on a low-power chip is improved.

The application discloses an echo cancellation method and device. The method comprises the following steps: acquiring an initial multi-channel audio signal to be subjected to echo cancellation; carrying out linear elimination processing on the obtained initial multi-channel audio signal to obtain a processed first residual multi-channel audio signal; carrying out nonlinear suppression processing on the processed first residual multi-channel audio signal by utilizing the masking effect of sound to obtain a second residual multi-channel audio signal. Therefore, the echo cancellation effect can be enhanced, the multi-channel audio signal is high in definition, and the effect of reducing scene echoes is achieved.

The invention discloses an echo cancellation method of a household appliance, the household appliance, a terminal and a storage medium. According to the method, based on an obtained first regulation and control instruction and/or a second regulation and control instruction input by a user, the household appliance is made to execute an echo cancellation function, echo cancellation error data outputby the household appliance are obtained, and a regulated and controlled echo cancellation optimization result is displayed. According to the technical scheme, a user can be supported to carry out field echo cancellation regulation and control on the household appliance in an interaction mode, the echo cancellation effect of the household appliance is improved by changing the size of the input voice signal and/or the output size of a self-adaptive filter, and the voice recognition accuracy and user experience of the household appliance are improved. The method can be widely applied to the technical field of household appliances.

A method for improving an echo cancellation effect and a system thereof are disclosed. The method comprises includes: performing a non-linear compensation for a non-linear response portion of an excitation signal using an NLC algorithm to obtain a compensated excitation signal; outputting the compensated excitation signal to an echo cancellation system; and performing echo cancellation for the compensated excitation signal by the echo cancellation system. According to the present disclosure, using the NLC algorithm, non-linear compensation is performed for the non-linear portion of the excitation signal, non-linear outputs generated due to non-linear characteristics of the system are pre-compensated when being input to the echo cancellation system, such that the echo signal output by the echo cancellation system is minimized and the echo cancellation effect is improved.