Voice signal filtering method, device, equipment, vehicle and program product

By presetting the filter parameters matching the adaptive filter with the microphone and speakers, the speaker echo is quickly eliminated, solving the problem of poor echo cancellation in the prior art, achieving clearer call quality and higher system adaptability.

CN119031306APending Publication Date: 2024-11-26CHONGQING CHANGAN TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411122557.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-11-26

Smart Images

  • Figure CN119031306A_ABST
    Figure CN119031306A_ABST
Patent Text Reader

Abstract

The invention relates to a voice signal filtering method, device and equipment, a vehicle and a program product, and relates to the technical field of voice processing. The method is applied to a vehicle, microphones and loudspeakers are arranged inside and outside the vehicle, and the method comprises the steps that a voice signal collected by a first microphone of the vehicle is acquired, and the voice signal comprises an audio signal sent by a first loudspeaker; the first microphone and the first loudspeaker are both located inside or outside a vehicle; a voice signal collected by a first microphone of the vehicle is filtered through a preset adaptive filter, a filtered voice signal is obtained, and filtering parameters of the preset adaptive filter are matched with the first microphone and the first loudspeaker; controlling a second loudspeaker to play the filtered voice signal; the space in which the second loudspeaker is arranged is different from the space in which the first loudspeaker is arranged. Therefore, the echo caused by the loudspeaker can be effectively eliminated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of speech processing technology, in particular to the field of echo cancellation technology, and specifically to a speech signal filtering method, device, equipment, vehicle and program product. Background Art

[0002] When making an audio call, the sound from the speaker may be captured by the microphone and transmitted to the other party, causing echo interference during the call. Echoes interfere with the clarity of the call and make it difficult for the other party to hear the content of the conversation. Echo cancellation technology can reduce or eliminate this interference and improve call quality.

[0003] However, the echo cancellation method in the related art is difficult to effectively cancel the echo caused by the speaker. Summary of the invention

[0004] The present application provides a method, device, equipment, vehicle and program product for filtering speech signals, so as to at least solve the technical problem that it is difficult to effectively eliminate the echo caused by the speaker in the related art. The technical solution of the present application is as follows:

[0005] According to a first aspect provided by the present application, a method for filtering a speech signal is provided, which is applied to a vehicle, wherein a microphone and a speaker are provided both inside and outside the vehicle, and comprises: obtaining a speech signal collected by a first microphone of the vehicle, wherein the speech signal comprises an audio signal emitted by a first speaker; wherein the first microphone and the first speaker are both located inside or outside the vehicle; filtering the speech signal collected by the first microphone of the vehicle by a preset adaptive filter to obtain a filtered speech signal, wherein the filtering parameters of the preset adaptive filter match the first microphone and the first speaker; and controlling the second speaker to play the filtered speech signal; wherein the second speaker is provided in a different space from the first speaker.

[0006] According to the above technical means, when the echo cancellation function is turned on, the present application can filter the audio signal emitted by the first speaker through the filtering parameters of the preset adaptive filter matched with the first microphone and the first speaker, accelerate the convergence speed of the preset adaptive filter, avoid the problem that the echo cannot be eliminated at the moment the echo cancellation function is turned on, and can more effectively filter out the echo generated by the audio signal emitted by the first speaker at the first microphone. Therefore, the present application can effectively eliminate the echo caused by the speaker.

[0007] In one possible embodiment, the method further includes: controlling the first speaker to play a preset audio signal, and controlling the first microphone to synchronously collect the audio signal played by the first speaker; the vehicle is located in a noise-free environment; and determining the filtering parameters based on the preset audio signal played by the first speaker and the audio signal collected by the first microphone.

[0008] According to the above technical means, the present application can determine the filtering parameters in a noise-free environment and eliminate the interference of external noise on the audio signal. In this way, the audio signal collected by the first microphone will more purely reflect the transmission characteristics of the audio signal played by the first speaker after passing through the internal or external acoustic environment of the vehicle. Therefore, the filtering parameters determined based on these signals will be more accurate and can more effectively eliminate echoes.

[0009] In one possible embodiment, the method further includes: obtaining a preset audio signal played by a first speaker of multiple vehicles of the same type as the vehicle, and an audio signal collected by a first microphone; for each vehicle, determining a transfer parameter based on the preset audio signal played by the first speaker of the vehicle, and the audio signal collected by the first microphone; determining the filtering parameter based on the preset audio signal played by the first speaker and the audio signal collected by the first microphone includes: determining an average value of multiple transfer parameters corresponding to the multiple vehicles one by one as the filtering parameter.

[0010] According to the above technical means, the present application can collect audio signals of multiple vehicles of the same type, and determine the transmission parameters based on the audio signals of multiple vehicles of the same type, so as to further calculate the average value thereof as the filtering parameter, so that the filtering parameter can be more representative and generalized. In this way, even when facing different individuals but vehicles of the same type, the filtering parameters can be well adapted, thereby improving the robustness of the system.

[0011] In one possible manner, the transfer parameter is determined based on a preset audio signal played by the first speaker of the vehicle and an audio signal collected by the first microphone, including: performing Fourier transform on the preset audio signal played by the first speaker of the vehicle to obtain a first frequency domain signal, and performing Fourier transform on the audio signal collected by the first microphone of the vehicle to obtain a second frequency domain signal; and determining the ratio between the absolute value of the first frequency domain signal and the absolute value of the second frequency domain signal as the transfer parameter.

[0012] According to the above-mentioned technical means, the present application can accurately determine the attenuation, phase change and other characteristics of the audio signal during the transmission process through the absolute value ratio of the first frequency domain signal (the frequency domain representation output by the speaker) and the second frequency domain signal (the frequency domain representation collected by the microphone), thereby obtaining accurate transmission parameters.

[0013] In one possible manner, the method further includes: acquiring a voice signal collected by a second microphone of the vehicle; the first microphone and the second microphone are both located inside or outside the vehicle; judging whether the current moment is in a dual-talk state based on the voice signal collected by the first microphone and the voice signal collected by the first microphone; the dual-talk state is used to characterize that when the first speaker plays an audio signal, other sound sources emit an audio signal; filtering the voice signal through a preset adaptive filter to obtain a filtered voice signal, including: when the current moment is not in a dual-talk state, updating the filtering parameters based on the voice data to obtain updated filtering parameters; filtering the voice signal based on a preset adaptive filter configured with the played audio signal.

[0014] According to the above technical means, the present application can more accurately determine whether the current state is a dual-talk state by comparing the voice signals collected simultaneously by the first microphone and the second microphone. This supports updating the filter parameters in a non-dual-talk state, ensuring that the filter can more accurately adapt to the acoustic characteristics of the current environment. By updating the filter parameters in real time, it can ensure that the filter always maintains the best state, thereby improving the accuracy of echo cancellation.

[0015] In one possible manner, the method further includes: filtering the voice signal through a preset adaptive filter to obtain a filtered voice signal, including: filtering the voice signal based on a preset adaptive filter that configures the filtering parameters when the current moment is a dual-talk state.

[0016] According to the above technical means, in the dual-talk state, in addition to the audio signal played by the speaker, there are other sound sources (such as passengers, external noise, etc.) that simultaneously send out audio signals. If the filter parameters are updated at this time, the audio signals from other sound sources may be mistakenly regarded as echo signals for elimination, resulting in voice distortion or loss. By keeping the filter parameters unchanged, it is possible to avoid the occurrence of mistakenly treating the audio signals from other sound sources as echo signals for elimination, ensuring the integrity and clarity of the voice signal.

[0017] In one possible manner, judging whether the current moment is in a dual-talk state based on the voice signal collected by the first microphone and the voice signal collected by the second microphone includes: performing feature extraction on the voice signal collected by the first microphone and the voice signal collected by the first microphone to obtain feature data; and judging whether the current moment is in a dual-talk state based on the feature data.

[0018] According to the above technical means, the present application can obtain more comprehensive acoustic environment information by extracting features from the voice signals collected by microphones at two different locations (such as different locations inside the vehicle). These feature data can reflect the distribution characteristics of different sound sources in time and space, so as to more accurately determine whether the current state is dual-speaking.

[0019] In one possible manner, speech enhancement is performed on the filtered speech signal to obtain an enhanced speech signal; and the second speaker is controlled to play the enhanced speech signal.

[0020] According to the above technical means, the present application can enhance the propagation distance and penetration of voice signals through voice enhancement technology, making communication smoother and more reliable.

[0021] According to a second aspect provided by the present application, a voice signal filtering device is provided, including: an acquisition unit, a processing unit and a control unit; the acquisition unit is used to acquire a voice signal collected by a first microphone of the vehicle, and the voice signal includes an audio signal emitted by a first speaker; the first microphone and the first speaker are both located inside or outside the vehicle; the processing unit is used to filter the voice signal collected by the first microphone of the vehicle through a preset adaptive filter to obtain a filtered voice signal, and the filtering parameters of the preset adaptive filter match the first microphone and the first speaker; the control unit is used to control the second speaker to play the filtered voice signal; the second speaker is set in a space different from that of the first speaker.

[0022] In one possible embodiment, the control unit is also used to control the first speaker to play a preset audio signal, and control the first microphone to synchronously collect the audio signal played by the first speaker; the vehicle is located in a noise-free environment; the processing unit is also used to determine the filtering parameters based on the preset audio signal played by the first speaker and the audio signal collected by the first microphone.

[0023] In one possible embodiment, the acquisition unit is further used to acquire preset audio signals played by the first speakers of multiple vehicles of the same type as the vehicle, and audio signals collected by the first microphone; the processing unit is further used to determine the transfer parameter for each vehicle based on the preset audio signal played by the first speaker of the vehicle, and the audio signal collected by the first microphone; the processing unit is specifically used to determine the average value of multiple transfer parameters corresponding to the multiple vehicles one by one as the filtering parameter.

[0024] In one possible embodiment, the processing unit is specifically used to: perform Fourier transform on a preset audio signal played by a first speaker of the vehicle to obtain a first frequency domain signal, and perform Fourier transform on an audio signal collected by a first microphone of the vehicle to obtain a second frequency domain signal; and determine the ratio between the absolute value of the first frequency domain signal and the absolute value of the second frequency domain signal as the transfer parameter.

[0025] In one possible manner, the acquisition unit is also used to acquire a voice signal collected by a second microphone of the vehicle; the first microphone and the second microphone are both located inside or outside the vehicle; the processing unit is used to determine whether the current moment is a dual-talk state based on the voice signal collected by the first microphone and the voice signal collected by the first microphone; the dual-talk state is used to characterize that when the first speaker plays an audio signal, other sound sources emit an audio signal; the processing unit is specifically used to: when the current moment is not a dual-talk state, update the filtering parameters based on the voice data to obtain updated filtering parameters; filter the voice signal based on a preset adaptive filter configured with the played audio signal.

[0026] In a possible manner, the processing unit is specifically configured to: when the current moment is a double-talk state, filter the speech signal based on a preset adaptive filter configured with the filtering parameters.

[0027] In one possible manner, the processing unit is specifically used to: perform feature extraction on the voice signal collected by the first microphone and the voice signal collected by the second microphone to obtain feature data; and determine whether the current moment is a dual-talk state based on the feature data.

[0028] In one possible embodiment, the processing unit is further used to perform speech enhancement on the filtered speech signal to obtain an enhanced speech signal; and the control unit is used to control the second speaker to play the enhanced speech signal.

[0029] According to the third aspect provided by the present application, an electronic device is provided, comprising: a processor; a memory for storing processor executable instructions; wherein the processor is configured to execute instructions to implement the method of the above-mentioned first aspect and any possible implementation manner thereof.

[0030] According to the fourth aspect provided by the present application, a computer-readable storage medium is provided. When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method in the above-mentioned first aspect and any possible implementation method thereof.

[0031] According to the fifth aspect provided by the present application, a computer program product is provided, the computer program product comprising computer instructions, and when the computer instructions are executed on an electronic device, the electronic device executes the method of the above-mentioned first aspect and any possible implementation manner thereof.

[0032] Therefore, the above technical features of the present application have the following beneficial effects:

[0033] (1) When the echo cancellation function is turned on, the audio signal emitted by the first speaker is filtered by the filter parameters of the preset adaptive filter matched with the first microphone and the first speaker, so as to accelerate the convergence speed of the preset adaptive filter, avoid the problem that the echo cannot be eliminated at the moment the echo cancellation function is turned on, and more effectively filter out the echo generated by the audio signal emitted by the first speaker at the first microphone. Therefore, the present application can effectively eliminate the echo caused by the speaker.

[0034] (2) Determining the filtering parameters in a noise-free environment can eliminate the interference of external noise on the audio signal. In this way, the audio signal collected by the first microphone will more purely reflect the transmission characteristics of the audio signal played by the first speaker after passing through the internal or external acoustic environment of the vehicle. Therefore, the filtering parameters determined based on these signals will be more accurate and can more effectively eliminate echoes.

[0035] (3) By collecting audio signals of multiple vehicles of the same type and determining the transmission parameters based on the audio signals of multiple vehicles of the same type, and further calculating the average value as the filtering parameter, the filtering parameter can be more representative and generalized. In this way, even when facing different individuals but the same type of vehicles, the filtering parameters can be better adapted, thereby improving the robustness of the system.

[0036] (4) By using the absolute value ratio of the first frequency domain signal (the frequency domain representation of the speaker output) and the second frequency domain signal (the frequency domain representation collected by the microphone), the attenuation, phase change and other characteristics of the audio signal during the transmission process can be accurately determined, thereby obtaining accurate transmission parameters.

[0037] (5) By comparing the voice signals collected by the first microphone and the second microphone at the same time, it is possible to more accurately determine whether the current state is a dual-talk state. This supports updating the filter parameters in a non-dual-talk state, ensuring that the filter can more accurately adapt to the acoustic characteristics of the current environment. By updating the filter parameters in real time, it is possible to ensure that the filter always maintains the best state, thereby improving the accuracy of echo cancellation.

[0038] (6) In the dual-talk state, in addition to the audio signal played by the speaker, there are other sound sources (such as passengers, external noise, etc.) that simultaneously send out audio signals. If the filter parameters are updated at this time, the audio signals from other sound sources may be mistakenly regarded as echo signals for elimination, resulting in speech distortion or loss. By keeping the filter parameters unchanged, the occurrence of mistakenly treating the audio signals from other sound sources as echo signals for elimination can be avoided, ensuring the integrity and clarity of the speech signal.

[0039] (7) By extracting features from the speech signals collected by microphones at two different locations (such as different locations inside the vehicle), more comprehensive acoustic environment information can be obtained. These feature data can reflect the distribution characteristics of different sound sources in time and space, thereby more accurately determining whether the current state is dual-speaking.

[0040] (8) Voice enhancement technology can enhance the propagation distance and penetration of voice signals, making communications smoother and more reliable.

[0041] It should be noted that the technical effects brought about by any implementation method in the second to fifth aspects can refer to the technical effects brought about by the corresponding implementation method in the first aspect, and will not be repeated here.

[0042] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the specification are used to explain the principles of the present application, and do not constitute an improper limitation on the present application.

[0044] Figure 1 is a structural schematic diagram of a speech signal filtering system according to an exemplary embodiment;

[0045] Figure 2 is a flow chart of a method for filtering a speech signal according to an exemplary embodiment;

[0046] Figure 3 is a schematic diagram of a vehicle according to an exemplary embodiment;

[0047] Figure 4 is a schematic diagram of a two-way real-time intercom process according to an exemplary embodiment;

[0048] Figure 5 is a schematic diagram of a vehicle-mounted speech framework according to an exemplary embodiment;

[0049] Figure 6is a block diagram of a speech signal filtering device according to an exemplary embodiment;

[0050] Figure 7 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0051] In order to enable ordinary persons in the art to better understand the technical solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0052] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims.

[0053] As described in the background technology, in order to solve the problem that it is difficult to effectively eliminate the echo caused by the speaker, the present application provides a method for filtering a voice signal, obtaining a voice signal collected by a first microphone of the vehicle, wherein the voice signal includes an audio signal emitted by a first speaker, and filtering the voice signal collected by the first microphone of the vehicle through a preset adaptive filter to obtain a filtered voice signal, thereby effectively eliminating the echo of the audio signal emitted by the first speaker, so as to further control the second speaker to play the filtered voice signal, that is, controlling the second speaker to play the voice signal after echo elimination.

[0054] For ease of understanding, the speech signal filtering method provided in the present application is specifically introduced below with reference to the accompanying drawings.

[0055] Figure 1 is a structural diagram of a speech signal filtering system according to an exemplary embodiment. Figure 1 As shown, the speech signal filtering system includes: a vehicle 101, a speech signal filtering device 102, a microphone 103, and a speaker 104. The speech signal filtering device 102, the microphone 103, and the speaker 104 may be configured in the vehicle 101. The speech signal filtering device 102 may be communicatively connected to the microphone 103. The speech signal filtering device 102 may be communicatively connected to the speaker 104.

[0056] For example, the connection may be made by wired or wireless means. The wireless means may include Bluetooth, wireless fidelity (WIFI), communication networks (such as the fifth generation (5th generation, 5G), etc. The wired means may include network cables, system buses, etc.

[0057] The number of microphones 103 may be one or more. Figure 1 Only one microphone 103 is shown.

[0058] The number of the loudspeaker 104 may be one or more. Figure 1 Only one speaker 104 is shown.

[0059] Figure 2 is a flow chart of a method for filtering a speech signal according to an exemplary embodiment. Figure 2 As shown, the voice signal filtering method includes the following steps: S201-S203.

[0060] S201. A voice signal filtering device obtains a voice signal collected by a first microphone of the vehicle.

[0061] The voice signal may include an audio signal emitted by a first speaker. The first microphone and the first speaker are both located inside or outside the vehicle.

[0062] In a possible manner, when the first speaker plays an audio signal, the voice signal filtering device can control the first microphone to collect the audio signal played by the first speaker, and the voice signal filtering device can obtain the voice signal collected by the first microphone of the vehicle.

[0063] In a possible manner, when the first microphone is a microphone inside the vehicle, the first speaker may be any microphone inside the vehicle. When the first microphone is a microphone outside the vehicle, the first speaker may be any microphone outside the vehicle.

[0064] S202: The voice signal filtering device filters the voice signal collected by the first microphone of the vehicle through a preset adaptive filter to obtain a filtered voice signal.

[0065] The filter parameters of the preset adaptive filter are matched with the first microphone and the first speaker.

[0066] In a possible manner, the voice signal filtering device may filter the voice signal collected by the first microphone of the vehicle by using a preset adaptive filter to obtain a filtered voice signal.

[0067] Optionally, the preset adaptive filter can be set according to actual needs. For example, the preset adaptive filter can be a non-recursive (finite impulse response, FIR) filter. This application does not make specific restrictions on this.

[0068] In one example, the preset adaptive filter can be a FIR filter, and the microphone and speaker inside the vehicle are taken as an example. The vector of the N-order FIR filter in the time domain can be expressed by the following formula 1:

[0069] w(n)=(w[0],w[1],...,w[N-1]) T Formula 1

[0070] Wherein, w(n) can be used to represent the vector of the N-order FIR filter. n can be used to represent the number of iterations. The FIR filter can be divided into p segments, each of which can be k in length. N is equal to p*k. The i-th segment filter can be represented by the following formula 2:

[0071]

[0072] Among them, w i (n) can be used to represent the vector representation of the i-th filter. p can be used to represent the number of filter segments. k can be used to represent the length of each filter segment.

[0073] The block representation of w(n) satisfies the following formula 3:

[0074]

[0075] Among them, w(n) can be used to represent the vector of the N-th order FIR filter. w0(n) can be used to represent the vector representation of the first stage filter. p-1 (n) can be used to characterize the vector representation of the p-th filter.

[0076] w i (n) is padded to 2k and converted to the frequency domain by Fourier transform to satisfy the following formula 4:

[0077]

[0078] Where F is used to represent the Fourier transform. i (n) can be used to characterize the vector representation of the i-th filter. Can be used to convert w i (n) is zero-filled to 2k and converted to a vector representation in the frequency domain by Fourier transform. k*1 Can be used to represent zero padding.

[0079] The vector representation of the speech signal x(n) collected by the microphone satisfies the following formula 5:

[0080]

[0081] Among them, x(n) is used to represent the vector representation of the speech signal collected by the microphone. x0(n) can be used to represent the first segment x(n). p-1 (n) can be used to characterize the pth segment x(n).

[0082] The new and old pieces of x(n) are concatenated and Fourier transformed to obtain the following formula 6:

[0083]

[0084] in, It can be used to represent the vector obtained by splicing the old and new pieces of x(n) and performing Fourier transform. i The length of (n) is k, which is used to represent the i-th segment of x(n). F is used to represent the Fourier transform.

[0085] The voice signal collected by the microphone is convolved with the filter in the time domain, that is, the product is performed in the frequency domain to obtain the actual output of the filter. The actual output of the filter satisfies the following formula 7:

[0086]

[0087] Among them, y(n) can be used to characterize the actual output of the filter. O is a vector of all 0s, I is a vector of all 1s, and F -1 is the inverse Fourier transform. 1*k , I 1*k ] can be used to characterize filtering parameters.

[0088] In one possible approach, the time domain error signal satisfies the following formula eight:

[0089] e(n)=d(n)-y(n) Formula 8

[0090] Among them, y(n) can be used to represent the actual output of the filter. d(n) can be used to represent the all-zero vector of length k. e(n) can be used to represent the time domain error signal.

[0091] In one possible manner, the voice signal filtering device can control the first speaker to play a preset audio signal and control the first microphone to synchronously collect the audio signal played by the first speaker, wherein the vehicle is located in a noise-free environment. The voice signal filtering device can determine the filtering parameters based on the preset audio signal played by the first speaker and the audio signal collected by the first microphone.

[0092] Optionally, the preset audio signal played by the first speaker may be a pink noise signal or a white noise signal, which is not specifically limited in the present application.

[0093] In one example, the voice signal filtering device can control the first speaker to play a pink noise signal or a white noise signal. The voice signal filtering device can control the first microphone to synchronously collect the pink noise signal or the white noise signal played by the first speaker. The voice signal filtering device can determine the filtering parameters based on the pink noise signal or the white noise signal played by the first speaker and the pink noise signal or the white noise signal collected by the first microphone.

[0094] In one possible manner, the filtering device for voice signals may obtain a preset audio signal played by the first speakers of multiple vehicles of the same type as the vehicle, and an audio signal collected by the first microphone. The filtering device for voice signals may determine the transfer parameter for each vehicle according to the preset audio signal played by the first speaker of the vehicle, and the audio signal collected by the first microphone. The filtering device for voice signals may determine the average value of multiple transfer parameters corresponding one by one to the multiple vehicles as the filter parameter. The specific implementation method of the filtering device for voice signals determining the filter parameter based on the preset audio signal played by the first speaker and the audio signal collected by the first microphone may refer to S301-S302 below. No further details will be given here.

[0095] In a possible manner, before filtering the voice signal collected by the first microphone of the vehicle through a preset adaptive filter, the voice signal filtering device needs to determine whether the current moment is a dual-talk state.

[0096] That is, the voice signal filtering device can obtain the voice signal collected by the second microphone of the vehicle. The voice signal filtering device can determine whether the current moment is a dual-talk state based on the voice signal collected by the first microphone and the voice signal collected by the first microphone. The specific implementation method of the voice signal filtering device determining whether the current moment is a dual-talk state can refer to the following S401-S402. It will not be repeated here.

[0097] In one possible manner, the voice signal filtering device can update the filtering parameters based on the voice data to obtain updated filtering parameters when the dual-talk state is not in the current moment. The voice signal filtering device can filter the voice signal based on a preset adaptive filter configured with the played audio signal.

[0098] In another possible manner, the voice signal filtering device may filter the voice signal based on a preset adaptive filter configured with the filtering parameters when the current moment is in a double-talk state.

[0099] S203: The voice signal filtering device controls the second speaker to play the filtered voice signal.

[0100] The second speaker is arranged in a space different from that of the first speaker.

[0101] In one example, when the second speaker is arranged outside the vehicle, the first speaker should be located inside the vehicle. When the second speaker is arranged inside the vehicle, the first speaker should be located outside the vehicle.

[0102] In a possible manner, the voice signal filtering device may perform voice enhancement on the filtered voice signal to obtain an enhanced voice signal. The voice signal filtering device may control the second speaker to play the enhanced voice signal.

[0103] In another possible manner, the voice signal filtering device can perform voice activity detection (VAD), beamforming, noise reduction, voice enhancement, sound effect processing and other processing operations on the filtered voice signal to obtain a processed voice signal. The voice signal filtering device can control the second speaker to play the processed voice signal.

[0104] In one example, the speech signal filtering device can be used to determine whether the filtered speech signal contains human voice or valid speech part based on VAD. The speech signal filtering device can perform beamforming and noise reduction on the filtered speech signal when it is determined that the filtered speech signal contains human voice or valid speech part.

[0105] Optionally, the method for performing beamforming can be set according to actual needs. For example, the method for performing beamforming can be a delay and beamforming method, a filtering and beamforming method, a generalized sidelobe cancellation beamforming method, and a minimum variance signal distortion-free response beamforming method. This application does not impose specific restrictions on this.

[0106] Optionally, the noise reduction algorithm for noise reduction can be set according to actual needs. For example, the noise reduction algorithm can be spectral subtraction, Wiener filtering, subspace method, and neural network-based method. This application does not make specific restrictions on this.

[0107] In one example, after performing beamforming and noise reduction on the filtered voice signal, the filtering device of the voice signal can perform sound effect processing on the filtered voice signal, and can instantly transform the speaker's voice into a variety of distinctive timbres, such as the futuristic "technological sound", the lively and cute "Minions sound", or the steady and heavy "uncle sound".

[0108] based on Figure 2According to the technical solution in the invention, when the echo cancellation function is turned on, the present application can filter the audio signal emitted by the first speaker through the filtering parameters of the preset adaptive filter matched with the first microphone and the first speaker, accelerate the convergence speed of the preset adaptive filter, avoid the problem that the echo cannot be eliminated at the moment the echo cancellation function is turned on, and can more effectively filter out the echo generated by the audio signal emitted by the first speaker at the first microphone. Therefore, the present application can effectively eliminate the echo caused by the speaker.

[0109] In some embodiments, in order to determine the filtering parameters based on the preset audio signal played by the first speaker and the audio signal collected by the first microphone, the voice signal filtering method provided in the embodiment of the present application also includes the following steps: S301-S303.

[0110] S301. A voice signal filtering device obtains preset audio signals played by first speakers of multiple vehicles of the same vehicle type and audio signals collected by a first microphone.

[0111] In one possible manner, the manner in which the voice signal filtering device obtains the preset audio signals played by the first speakers of multiple vehicles of the same vehicle type and the audio signals collected by the first microphone can refer to the manner in which the voice signal filtering device obtains the voice signals collected by the first microphone of the vehicle in S201. No further details will be given here.

[0112] In one example, for each of the multiple vehicles of the same type as the vehicle, the voice signal filtering device can obtain the audio signal x_om1 collected by the first microphone and the preset audio signal y_or1 played by the first speaker.

[0113] S302: The voice signal filtering device determines a transfer parameter for each vehicle according to a preset audio signal played by a first speaker of the vehicle and an audio signal collected by a first microphone.

[0114] In one possible manner, the voice signal filtering device may perform Fourier transform on the preset audio signal played by the first speaker of the vehicle to obtain a first frequency domain signal, and perform Fourier transform on the audio signal collected by the first microphone of the vehicle to obtain a second frequency domain signal. The voice signal filtering device may determine the ratio between the absolute value of the first frequency domain signal and the absolute value of the second frequency domain signal as the transfer parameter.

[0115] In one example, the voice signal filtering device can perform Fourier transform on the audio signal x_om1 collected by the first microphone to obtain a first frequency domain signal X_om1, and the voice signal filtering device can perform Fourier transform on the preset audio signal y_or1 played by the first speaker to obtain a second frequency domain signal Y_om1.

[0116] Among them, the number of Fourier transform points N is 256, 512, 1024, etc.

[0117] The speech signal filtering device can calculate the absolute values ​​of the first frequency domain signal X_om1 and the second frequency domain signal Y_om1 respectively, and determine the ratio between the absolute value of the first frequency domain signal and the absolute value of the second frequency domain signal as the transfer parameter P_Cx1. The transfer parameter satisfies the following formula 9:

[0118]

[0119] Wherein, P_Cx1 can be used to characterize the transfer parameter of vehicle 1. X_om1(k) can be used to characterize the first frequency domain signal. Y_om1(k) can be used to characterize the second frequency domain signal.

[0120] S303. The voice signal filtering device determines the average value of the multiple transfer parameters corresponding to the multiple vehicles as the filtering parameter.

[0121] In a possible manner, the speech signal filtering device may determine the average value of a plurality of transfer parameters corresponding one-to-one to the plurality of vehicles as the filtering parameter.

[0122] In one example, the filtering parameter P satisfies the following tenth formula:

[0123]

[0124] Wherein, P can be used to characterize the filtering parameters. P_Cx1 can be used to characterize the transfer parameters of vehicle 1. P_Cx2 can be used to characterize the transfer parameters of vehicle 2. P_Cxn can be used to characterize the transfer parameters of vehicle n. n can be used to characterize the number of vehicles.

[0125] In one possible manner, the speech signal filtering device performs Fourier transform on the filtering parameters to the time domain to obtain the filtering parameters in the time domain.

[0126] Based on the above technical solution, the present application can collect audio signals of multiple vehicles of the same type, and determine the transmission parameters based on the audio signals of multiple vehicles of the same type, so as to further calculate the average value thereof as the filtering parameter, so that the filtering parameter can be more representative and generalized. In this way, even when facing different individuals but vehicles of the same type, the filtering parameters can be well adapted, thereby improving the robustness of the system.

[0127] In some embodiments, the speech signal filtering method provided in the embodiments of the present application further includes the following steps: S401-S402.

[0128] S401. A voice signal filtering device obtains a voice signal collected by a second microphone of the vehicle.

[0129] The first microphone and the second microphone are both located inside or outside the vehicle.

[0130] In one example, when the first microphone is disposed outside the vehicle, the first microphone should be located outside the vehicle. When the first microphone is disposed inside the vehicle, the second microphone should be located inside the vehicle.

[0131] In one possible manner, the specific implementation manner of the voice signal filtering device acquiring the voice signal collected by the second microphone of the vehicle can refer to the implementation manner of the voice signal filtering device acquiring the voice signal collected by the first microphone of the vehicle in S201, which will not be described in detail here.

[0132] Optionally, the number of the second microphones can be set according to actual needs. For example, the number of the second microphones can be two or three. This application does not impose any specific restrictions on this.

[0133] S402: The voice signal filtering device determines whether the current moment is in a dual-talk state based on the voice signal collected by the first microphone and the voice signal collected by the second microphone.

[0134] The dual-talk state is used to indicate that when the first speaker plays an audio signal, other sound sources send out audio signals.

[0135] In one example, taking the number of the second microphones as two, the second microphone 1 and the second microphone 2 are respectively arranged at the rearview mirror positions of the vehicle.

[0136] The voice signal filtering device obtains a voice signal x_om2 collected by the second microphone 1 of the vehicle and a voice signal x_om3 collected by the second microphone 2 of the vehicle, as well as a voice signal x_om1 collected by the first microphone and an audio signal y_om1 played by the first speaker.

[0137] The filtering device of the speech signal can calculate parameter 1 and parameter 2 based on x_om1, and parameter 1 satisfies the following formula 11:

[0138]

[0139] Wherein, ρ1 is used to represent parameter 1. x_om1 can be used to represent the speech signal collected by the first microphone. y_om1 can be used to represent the audio played by the first speaker.

[0140] Parameter 2 satisfies the following formula twelve:

[0141]

[0142] Wherein, ρ2 is used to characterize parameter 2. x_om1 can be used to characterize the speech signal collected by the first microphone. E can be used to characterize the residual.

[0143] In (1-ρ1 2 )>0.6&ρ2 2 >0.92, it is determined that the double-talk condition 1 is satisfied, and the feature 1 D1(l) of the l-th frame is determined to be 1; otherwise, the feature 1 D1(l) of the l-th frame is determined to be 0.

[0144] The filtering device of the speech signal can calculate parameter 3 based on x_om1 and y_om1, and parameter 3 satisfies the following formula 13:

[0145]

[0146] Wherein, thr(l) can be used to characterize parameter 3 of the lth frame. x_om1 can be used to characterize the speech signal collected by the first microphone. y_om1 can be used to characterize the audio played by the first speaker.

[0147] When the parameter 3 of each frame satisfies the first preset condition, the feature 2 D2(1) of the lth frame is determined to be 1; otherwise, the feature 2 D2(1) of the lth frame is determined to be 0.

[0148] The speech signal filtering device can determine the feature three D3(l) of the lth frame to be 1 when the following formula 14 is satisfied between the x_om1, x_om2, and x_om3 signals of the lth frame, and determine the feature one D3(l) of the lth frame to be 0 otherwise.

[0149] Formula 14:

[0150]

[0151] Among them, x_om1 can be used to characterize the voice signal collected by the first microphone. x_om2 can be used to characterize the voice signal collected by the second microphone 1. x_om3 can be used to characterize the voice signal collected by the second microphone 2. β1 can be used to characterize the second preset condition. Can be used to characterize the third preset condition.

[0152] For each frame signal, feature 1 D1, feature 2 D2 and feature 3 D3 are smoothed using smoothing factors λ1, λ2 and λ3 respectively to obtain Dx1, Dx2 and Dx3, whose formulas are as follows: Formula 15, Formula 16 and Formula 17:

[0153] Dx1(l)=λ1*Dx1(l-1)+(1-λ1)*D1(l) Formula 15

[0154] Dx2(l)=λ2*Dx2(l-1)+(1-λ2)*D2(l) Formula 16

[0155] Dx3(l)=λ3*Dx3(l-1)+(1-λ3)*D3(l) Formula 17

[0156] The filtering device of the speech signal can obtain the following judgment parameters for Dx1, Dx2, and Dx3 obtained for each frame signal, and the judgment parameters satisfy the following formula 18:

[0157] Dx(l)=η1Dx1(l)+η2Dx2(x)+η3Dx3(x) Formula 18

[0158] Wherein, Dx(l) can be used to characterize the judgment parameter of the lth frame signal. η1+η2+η3=1. η1 can be used to characterize the weight of DX1(l). η2 can be used to characterize the weight of DX2(l). η3 can be used to characterize the weight of DX3(l).

[0159] The speech signal filtering device may determine the state of the lth frame signal as a double-talk state when Dx(l) is greater than a preset threshold.

[0160] Based on this, the present application can more accurately determine whether the current state is a dual-talk state by comparing the voice signals collected simultaneously by the first microphone and the second microphone. This supports updating the filter parameters in a non-dual-talk state, ensuring that the filter can more accurately adapt to the acoustic characteristics of the current environment. By updating the filter parameters in real time, it can ensure that the filter always maintains the best state, thereby improving the accuracy of echo cancellation.

[0161] In some embodiments, Figure 3 Shown is a schematic diagram of a vehicle.

[0162] In one possible way, Figure 3 The vehicle in the embodiment includes a microphone outside the vehicle, a speaker outside the vehicle, a microphone inside the vehicle, and a speaker inside the vehicle.

[0163] In some embodiments, in combination Figure 3 ,like Figure 4 The figure shows a schematic diagram of a two-way real-time intercom process.

[0164] In one possible manner, the voice signal filtering device can control the microphone inside the vehicle to collect the voice signal. The voice signal filtering device can perform echo cancellation on the voice signal collected by the microphone when the speaker inside the vehicle emits an audio signal. The voice signal filtering device can perform noise processing and sound effect processing on the voice signal when the speaker inside the vehicle does not emit an audio signal or after the echo cancellation processing, and control the speaker outside the vehicle to play the processed voice signal.

[0165] The voice signal filtering device can control the microphone outside the vehicle to collect voice signals. The voice signal filtering device can perform echo cancellation on the voice signal collected by the microphone when the external speaker of the vehicle emits an audio signal. The voice signal filtering device can perform voice enhancement processing on the voice signal when the external speaker of the vehicle does not emit an audio signal or after the echo cancellation processing, and control the speaker inside the vehicle to play the processed voice signal.

[0166] Optionally, the echo cancellation methods inside the vehicle and outside the vehicle may be the same or different, and this application does not impose any specific restrictions on this.

[0167] Based on this, when driving late at night, if the vehicle ahead blocks the road for a long time, the user can safely use the car's external voice system to directly shout and ask about the situation without risking opening the window or door, ensuring personal safety. Or, when driving in the cold winter, the user can conveniently use the car's external communication system to ask the passers-by outside the window for directions without opening the window, thus keeping the car warm and comfortable.

[0168] In some embodiments, Figure 5 , which is a schematic diagram of an in-vehicle voice framework.

[0169] In one possible embodiment, the in-vehicle voice framework includes a microphone array inside the vehicle, a speaker inside the vehicle, a microphone array outside the vehicle, a speaker outside the vehicle, an echo cancellation module A, a VAD module, a beamforming module, a noise reduction module, a gain control module, a sound effect processing module A, an echo cancellation module B, and a sound effect processing module B.

[0170] The microphone array inside the vehicle is used to collect in-vehicle voice, in-vehicle noise, and sound signals emitted by the speakers inside the vehicle.

[0171] The echo cancellation module A can perform echo cancellation on the sound signal collected by the microphone array inside the vehicle. The echo cancellation module includes a preset filter. When the microphone array inside the vehicle has M microphones and N speakers, the number of channels is M*N.

[0172] The VAD module can be used to perform voice detection on the voice signal. The beamforming module can be used to perform beamforming on the voice signal. The noise reduction module can be used to perform noise reduction on the voice signal. The gain control module can be used to perform gain control on the voice signal. The sound effect processing module A can be used to perform sound effect processing on the voice signal.

[0173] The microphone array outside the vehicle is used to collect voices outside the vehicle, noise outside the vehicle, and sound signals emitted by speakers outside the vehicle.

[0174] The echo cancellation module B can perform echo cancellation on the sound signal collected by the microphone array inside the vehicle. The echo cancellation module includes a preset filter. Among them, the echo cancellation module A can be the same or different. This application does not make specific restrictions on this.

[0175] The above mainly introduces the solution provided by the embodiment of the present application from the perspective of the method. In order to achieve the above functions, the filtering device or electronic device of the speech signal includes a hardware structure and / or software module corresponding to the execution of each function. It should be easily appreciated by those skilled in the art that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0176] The embodiment of the present application can divide the functional modules of the filtering device or electronic device for speech signals according to the above method. For example, the filtering device or electronic device for speech signals can include various functional modules corresponding to the various functional divisions, or two or more functions can be integrated into one processing module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the embodiment of the present application is schematic and is only a logical functional division. There may be other division methods in actual implementation.

[0177] Figure 6 is a block diagram of a speech signal filtering device according to an exemplary embodiment. Figure 6 The speech signal filtering device includes: an acquisition unit 501, a processing unit 502 and a control unit 503.

[0178] In a possible manner, the acquisition unit 501 is used to acquire a voice signal collected by a first microphone of the vehicle, where the voice signal includes an audio signal emitted by a first speaker.

[0179] In one possible embodiment, the processing unit 502 is used to filter the voice signal collected by the first microphone of the vehicle through a preset adaptive filter to obtain a filtered voice signal, and the filtering parameters of the preset adaptive filter match the first microphone and the first speaker.

[0180] In a possible manner, the control unit 503 is used to control a second speaker to play the filtered voice signal; the second speaker is set in a space different from that of the first speaker.

[0181] In a possible manner, the control unit 503 is further configured to control the first speaker to play a preset audio signal, and control the first microphone to synchronously collect the audio signal played by the first speaker.

[0182] In a possible manner, the processing unit 502 is further configured to determine the filtering parameter based on a preset audio signal played by the first speaker and an audio signal collected by the first microphone.

[0183] In a possible manner, the acquisition unit 501 is further used to acquire preset audio signals played by first speakers of multiple vehicles of the same type as the vehicle, and audio signals collected by the first microphone.

[0184] In a possible manner, the processing unit 502 is further configured to determine, for each vehicle, a transfer parameter based on a preset audio signal played by a first speaker of the vehicle and an audio signal collected by a first microphone.

[0185] In one possible manner, the processing unit is specifically configured to determine an average value of a plurality of transfer parameters corresponding one-to-one to the plurality of vehicles as the filtering parameter.

[0186] In one possible embodiment, the processing unit 502 is specifically used to: perform Fourier transform on a preset audio signal played by a first speaker of the vehicle to obtain a first frequency domain signal, and perform Fourier transform on an audio signal collected by a first microphone of the vehicle to obtain a second frequency domain signal; and determine the ratio between the absolute value of the first frequency domain signal and the absolute value of the second frequency domain signal as the transfer parameter.

[0187] In a possible manner, the acquisition unit 501 is further used to acquire a voice signal collected by a second microphone of the vehicle; the first microphone and the second microphone are both located inside or outside the vehicle.

[0188] In one possible manner, the processing unit 502 is used to determine whether the current moment is a dual-talk state based on the voice signal collected by the first microphone and the voice signal collected by the second microphone; the dual-talk state is used to characterize that when the first speaker plays the audio signal, other sound sources emit audio signals.

[0189] In one possible manner, the processing unit 502 is specifically used to: when the current moment is not a dual-talk state, update the filtering parameters based on the voice data to obtain updated filtering parameters; and filter the voice signal based on a preset adaptive filter configured with the played audio signal.

[0190] In a possible manner, the processing unit 502 is specifically configured to: when the current moment is a double-talk state, filter the speech signal based on a preset adaptive filter configured with the filtering parameters.

[0191] In one possible manner, the processing unit 502 is specifically used to: perform feature extraction on the voice signal collected by the first microphone and the voice signal collected by the second microphone to obtain feature data; and determine whether the current moment is a dual-talk state based on the feature data.

[0192] In a possible manner, the processing unit 502 is further configured to perform speech enhancement on the filtered speech signal to obtain an enhanced speech signal.

[0193] In a possible manner, the control unit 503 is used to control the second speaker to play the enhanced voice signal.

[0194] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0195] Figure 7 FIG. 1 is a block diagram of an electronic device according to an exemplary embodiment. Figure 7 As shown, the electronic device includes but is not limited to: a processor 601 and a memory 602 .

[0196] The memory 602 is used to store executable instructions of the processor 601. It can be understood that the processor 601 is configured to execute instructions to implement the voice signal filtering method in the above embodiment.

[0197] It should be noted that those skilled in the art can understand that Figure 7 The electronic device structure shown in the figure does not constitute a limitation on the electronic device, and the electronic device may include Figure 7More or fewer components may be shown, or certain components may be combined, or the components may be arranged differently.

[0198] The processor 601 is the control center of the electronic device. It uses various interfaces and lines to connect various parts of the entire electronic device. By running or executing software programs and / or modules stored in the memory 602, and calling data stored in the memory 602, it performs various functions of the electronic device and processes data, thereby monitoring the electronic device as a whole. The processor 601 may include one or more processing units. Optionally, the processor 601 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 601.

[0199] The memory 602 may be used to store software programs and various data. The memory 602 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required by at least one functional module (such as a determination unit, a processing unit, etc.), etc. In addition, the memory 602 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0200] In an exemplary embodiment, a computer-readable storage medium including instructions is also provided, such as a memory 602 including instructions, and the above instructions can be executed by a processor 601 of an electronic device to implement the method in the above embodiment.

[0201] In actual implementation, Figure 6 The functions of the acquisition unit 501, the processing unit 502 and the control unit 503 in the embodiment can be represented by Figure 7 The processor 601 in the embodiment calls the computer program stored in the memory 602. The specific execution process can refer to the description of the method part in the above embodiment, which will not be repeated here.

[0202] Optionally, the computer-readable storage medium may be a non-temporary computer-readable storage medium, for example, the non-temporary computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0203] In an exemplary embodiment, the present application also provides a computer program product including one or more instructions, and the one or more instructions can be executed by the processor 601 of the electronic device to complete the method in the above embodiment.

[0204] It should be noted that when the instructions in the above-mentioned computer-readable storage medium or one or more instructions in the computer program product are executed by the processor of the electronic device, the various processes of the above-mentioned method embodiment are implemented, and the same technical effect as the above-mentioned method can be achieved. To avoid repetition, they will not be repeated here.

[0205] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0206] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0207] The units described as separate components may or may not be physically separated, and the components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0208] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0209] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or the full classification part or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium, including a number of instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) to execute the full classification part or part of the steps of each embodiment method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, ROM, RAM, disk or CD and other media that can store program code.

[0210] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto, and any changes or substitutions within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for filtering a speech signal, characterized in that: Applied to a vehicle, the vehicle is provided with a microphone and a speaker both inside and outside, including: Acquire a voice signal collected by a first microphone of the vehicle, wherein the voice signal includes an audio signal emitted by a first speaker; the first microphone and the first speaker are both located inside or outside the vehicle; Filtering the voice signal collected by the first microphone of the vehicle by a preset adaptive filter to obtain a filtered voice signal, wherein the filtering parameters of the preset adaptive filter match the first microphone and the first speaker; A second speaker is controlled to play the filtered voice signal; the second speaker is set in a space different from that of the first speaker.

2. The method according to claim 1, characterized in that The method further comprises: Controlling the first speaker to play a preset audio signal, and controlling the first microphone to synchronously collect the audio signal played by the first speaker; the vehicle is located in a noise-free environment; The filtering parameter is determined based on a preset audio signal played by the first speaker and an audio signal collected by the first microphone.

3. The method according to claim 2, characterized in that The method further comprises: Acquire preset audio signals played by first speakers of a plurality of vehicles of the same type as the vehicle, and audio signals collected by a first microphone; For each vehicle, determining a transfer parameter according to a preset audio signal played by a first speaker of the vehicle and an audio signal collected by a first microphone; The determining the filtering parameter based on the preset audio signal played by the first speaker and the audio signal collected by the first microphone includes: An average value of a plurality of transfer parameters corresponding one to one to the plurality of vehicles is determined as the filtering parameter.

4. The method according to claim 3, characterized in that The determining of the transfer parameter according to the preset audio signal played by the first speaker of the vehicle and the audio signal collected by the first microphone includes: Performing Fourier transform on a preset audio signal played by a first speaker of the vehicle to obtain a first frequency domain signal, and performing Fourier transform on an audio signal collected by a first microphone of the vehicle to obtain a second frequency domain signal; The ratio between the absolute value of the first frequency domain signal and the absolute value of the second frequency domain signal is determined as the transfer parameter.

5. The method according to any one of claims 1 to 4, characterized in that The method further comprises: Acquire a voice signal collected by a second microphone of the vehicle; the first microphone and the second microphone are both located inside or outside the vehicle; Based on the voice signal collected by the first microphone and the voice signal collected by the second microphone, determining whether the current moment is in a dual-talk state; the dual-talk state is used to indicate that when the first speaker plays the audio signal, other sound sources send out audio signals; The filtering of the speech signal by a preset adaptive filter to obtain a filtered speech signal includes: When the dual-talk state is not in the current moment, updating the filter parameters based on the voice data to obtain updated filter parameters; The speech signal is filtered based on a preset adaptive filter configured with the played audio signal.

6. The method according to claim 5, characterized in that The filtering of the speech signal by a preset adaptive filter to obtain a filtered speech signal includes: In the case of a double talk state at the current moment, the speech signal is filtered based on a preset adaptive filter configured with the filtering parameters.

7. The method according to claim 5, characterized in that The determining whether the current moment is a dual-talk state based on the voice signal collected by the first microphone and the voice signal collected by the second microphone includes: Extracting features of the speech signal collected by the first microphone and the speech signal collected by the second microphone to obtain feature data; Based on the characteristic data, it is determined whether the current moment is a dual talk state.

8. The method according to claim 1, characterized in that The method further comprises: Performing speech enhancement on the filtered speech signal to obtain an enhanced speech signal; The second speaker is controlled to play the enhanced voice signal.

9. A speech signal filtering device, characterized in that: The device comprises: an acquisition unit, a processing unit and a control unit; The acquisition unit is used to acquire a voice signal collected by a first microphone of the vehicle, wherein the voice signal includes an audio signal emitted by a first speaker; the first microphone and the first speaker are both located inside or outside the vehicle; The processing unit is used to filter the voice signal collected by the first microphone of the vehicle through a preset adaptive filter to obtain a filtered voice signal, wherein the filtering parameters of the preset adaptive filter match the first microphone and the first speaker; The control unit is used to control the second speaker to play the filtered voice signal; the second speaker is set in a space different from that of the first speaker.

10. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the method according to any one of claims 1 to 8.

11. A vehicle, characterized in that: The vehicle comprises the device of claim 9.

12. A computer program product comprising instructions, characterized in that When the instructions are executed by a computer, the computer is caused to perform the method according to any one of claims 1 to 8.