Voice signal processing method, apparatus, device, medium, and computer program product

Abstract

Embodiments of the present application disclose a voice signal processing method, device, equipment, medium and computer program product, wherein the method comprises: acquiring a first voice signal collected by a first terminal, a second voice signal transmitted by the first terminal to a second terminal, and a third voice signal transmitted by the second terminal to the first terminal; performing adaptive filtering processing on the second voice signal and the third voice signal based on an initial plurality of adaptive filters respectively, to obtain interference signals in the simulated first voice signal; and performing suppression processing on the interference signals in the first voice signal to obtain a target voice signal after suppression processing. The present application can realize the effect of suppressing howling in the process of communication, effectively optimize the short-distance communication effect, avoid the obvious distortion of the reverberation tail caused by howling, and thus effectively improve the voice quality of communication.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a speech signal processing method, apparatus, device, medium, and computer program product. Background Technology

[0002] With the continuous development of voice signal processing technology, users have increasingly higher requirements for voice quality. However, echoes or feedback are prone to occur in daily calls, seriously affecting voice quality, especially during short-distance calls. The signal from the microphone at the near end is received by the speaker at the near end through the acoustic path, and then returns from the network sidetone or the acoustic path of the other end, thus forming feedback. If this feedback becomes positive feedback, it will cause feedback. Feedback has a significant impact on the microphone. For example, the delayed feedback of the speaker's sound field will cause the entire system to form a series of delayed echoes, and these echoes will aggravate the comb filtering effect, producing obvious distorted reverberation tails, seriously affecting the quality of call voice. Summary of the Invention

[0003] In view of this, this application provides a voice signal processing method, apparatus, medium, and computer program product to solve the problem of call quality degradation caused by howling in existing call technologies. To achieve one or more of the above objectives, or other objectives, this application proposes a voice signal processing method, apparatus, medium, and computer program product.

[0004] First aspect: A speech signal processing method, comprising:

[0005] Acquire a first voice signal collected by the first terminal, a second voice signal transmitted from the first terminal to the second terminal, and a third voice signal transmitted from the second terminal to the first terminal;

[0006] Based on the initial multiple adaptive filters, the second speech signal and the third speech signal are adaptively filtered respectively to obtain the interference signal in the simulated first speech signal; the interference signal is used to update the adaptive filter, and the updated adaptive filter is used to continue filtering the second speech signal and the third speech signal in subsequent frames;

[0007] The interference signal in the first speech signal is suppressed to obtain the suppressed target speech signal.

[0008] Second aspect: A speech signal processing device, comprising:

[0009] The acquisition module is used to acquire a first voice signal collected by the first terminal, a second voice signal transmitted from the first terminal to the second terminal, and a third voice signal transmitted from the second terminal to the first terminal;

[0010] A filtering module is used to adaptively filter the second speech signal and the third speech signal based on an initial set of multiple adaptive filters to obtain an interference signal in the simulated first speech signal; the interference signal is used to update the adaptive filters, and the updated adaptive filters are used to continue filtering the second speech signal and the third speech signal in subsequent frames;

[0011] The suppression module is used to suppress interference signals in the first speech signal to obtain the suppressed target speech signal.

[0012] In one embodiment, the first voice signal includes a voice signal emitted by a first user, a voice signal corresponding to a first interference path between the first terminal and the second terminal, and a voice signal corresponding to a second interference path of the first terminal itself.

[0013] In one embodiment, the first interference path is the path for transmitting voice signals through space between the first terminal and the second terminal; the second interference path is the path for transmitting voice signals through space between the first terminal and the first terminal.

[0014] In one embodiment, the second voice signal includes the voice signal transmitted from the first terminal to the second terminal via the target path; the third voice signal includes the voice signal transmitted from the second terminal to the first terminal via the target path.

[0015] In one embodiment, the adaptive filter includes a first adaptive filter and a second adaptive filter; the interference signal includes a first interference signal and a second interference signal.

[0016] The filtering module includes:

[0017] The first filtering unit is used to filter the second speech signal through the first adaptive filter to obtain the first interference signal in the simulated first speech signal corresponding to the first interference path.

[0018] The second filtering unit is used to filter the third speech signal through the second adaptive filter to obtain the second interference signal in the simulated first speech signal corresponding to the second interference path.

[0019] In one embodiment, the first filtering unit is specifically used to filter the second speech signal according to the current filter parameters using the first adaptive filter to obtain a first interference signal in the simulated first speech signal that corresponds to the first interference path; the first interference signal is an analog signal that has similar characteristics to the second speech signal.

[0020] In one embodiment, the second filtering unit is specifically used to filter the third speech signal according to the current filter parameters using the second adaptive filter to obtain a second interference signal in the simulated first speech signal that corresponds to the second interference path; the second interference signal is an analog signal that has similar characteristics to the third speech signal.

[0021] In one embodiment, the apparatus further includes an update module, the update module comprising:

[0022] The first update unit is used to update the first adaptive filter corresponding to the first interference path based on the first interference signal and the voice signal corresponding to the first interference path between the first terminal and the second terminal.

[0023] The second update unit is used to update the second adaptive filter corresponding to the second interference path based on the second interference signal and the voice signal corresponding to the second interference path of the first terminal itself.

[0024] In one embodiment, the interference signal includes a first interference signal and a second interference signal; the suppression module is specifically used to: perform suppression processing on the first interference signal and the second interference signal in the first speech signal to remove the first interference signal and the second interference signal from the first speech signal to obtain the suppressed target speech signal.

[0025] Third aspect: An electronic device, comprising at least one processor and at least one memory, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the speech signal processing methods of the embodiments of this application.

[0026] Fourth aspect: A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the speech signal processing methods of the embodiments of this application.

[0027] Fifth aspect: A computer program product, comprising a computer program that, when executed by a processor, implements the steps of the speech signal processing methods of the embodiments of this application.

[0028] Implementing the embodiments of this application will have the following beneficial effects:

[0029] First, the first voice signal acquired by the first terminal, the second voice signal transmitted from the first terminal to the second terminal, and the third voice signal transmitted from the second terminal to the first terminal are acquired. Then, based on initial multiple adaptive filters, adaptive filtering processing is performed on the second and third voice signals respectively to obtain interference signals in the simulated first voice signal. The adaptive filters are then updated using the interference signals, and the updated adaptive filters are used to continue filtering the second and third voice signals in subsequent frames. Finally, the interference signals in the first voice signal are suppressed to obtain the suppressed target voice signal. This application adds multiple adaptive filters to the full-duplex communication system and adopts a feedback cancellation approach. During full-duplex communication, adaptive filters at both near and far ends are simultaneously invoked to predict the actual feedback path and match it with the adaptive filters, thereby cutting off the feedback path and suppressing howling during the call. This effectively optimizes short-distance call performance, avoids significant reverberation tails caused by howling, and thus effectively improves call voice quality. Attached Figure Description

[0030] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] in:

[0032] Figure 1 This is an application environment diagram of the speech signal processing method provided in the embodiments of this application;

[0033] Figure 2 A schematic flowchart illustrating the speech signal processing method provided in an embodiment of this application;

[0034] Figure 3 A schematic diagram of a near-field full-duplex call scenario provided in an embodiment of this application;

[0035] Figure 4 A schematic diagram illustrating one embodiment of the speech signal processing method provided in this application.

[0036] Figure 5 A schematic diagram illustrating another embodiment of the speech signal processing method provided in this application;

[0037] Figure 6 A schematic diagram of a communication scenario between an inner door screen and an outer door peephole provided in an embodiment of this application;

[0038] Figure 7This is a schematic diagram of the structure of the speech signal processing device provided in the embodiments of this application;

[0039] Figure 8 This is an internal structural diagram of a computer device provided in an embodiment of this application. Detailed Implementation

[0040] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0041] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0042] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0043] The speech signal processing method provided in this application can be applied to, for example... Figure 1 The application environment shown. Among them, Figure 1 A speech signal processing system is provided. The target detection system includes a smart device 100, a network device 200 communicatively connected to the smart device 100, a server 300, and a user terminal 400. The smart device 100 may include multiple devices, such as smart device 1002 and smart device 1004.

[0044] The smart device 100 connects to the network device 200 in the target detection system and communicates with the network device 200 through its own configured communication module. In one embodiment, the smart device 100 connects to the network device 200 via a local area network (LAN) or a wide area network (WAN), thereby being deployed within the network device 200. The LAN may include ZigBee or Bluetooth, and the WAN may include 2G / 3G / 4G / 5G / Wi-Fi, etc.

[0045] Network device 200 can establish network connections with server 300 and user terminal 400. In one embodiment, network device 200 and user terminal 400 can establish network connections through a local area network or wide area network path. Through this network connection, the user interacts with user terminal 400, thereby enabling the user to control the smart device 100 connected to network device 200 to perform corresponding actions via user terminal 400 or smart device 100.

[0046] In the specific implementation process, taking the first terminal as smart device 1002 and the second terminal as smart device 1004 as an example, the system acquires the first voice signal collected by smart device 1002, the second voice signal transmitted from smart device 1002 to smart device 1004, and the third voice signal transmitted from smart device 1004 to smart device 1002. Based on the initial multiple adaptive filters, the second and third voice signals are adaptively filtered to obtain the interference signal in the simulated first voice signal. The interference signal is used to update the adaptive filters, and the updated adaptive filters are used to continue filtering the second and third voice signals in subsequent frames. The interference signal in the first voice signal is suppressed, thereby smart device 1002 can obtain the suppressed target voice signal.

[0047] Furthermore, this application can also be implemented through hardware circuits or hardware circuits combined with software instructions. Therefore, the implementation of this application is not limited to any specific hardware circuit, software, or combination thereof.

[0048] The following sections provide detailed descriptions of each example. It should be noted that the order in which the embodiments are described is not intended to limit the priority of the embodiments.

[0049] Please see Figure 2 This paper provides a speech signal processing method, and takes its application in an electronic device as an example for illustration. Specifically, the electronic device may be... Figure 1 The application does not limit the scope to smart devices, terminals, gateways, servers, etc. The voice signal processing method specifically includes the following steps:

[0050] S1. Acquire the first voice signal collected by the first terminal, the second voice signal transmitted from the first terminal to the second terminal, and the third voice signal transmitted from the second terminal to the first terminal;

[0051] The first terminal is a device with communication capabilities that can directly interact with the user; specifically, it can be a communication device including a sound acquisition module and a sound playback module. The second terminal is also a device with communication capabilities that can directly interact with the user; specifically, it can be a communication device including a sound acquisition module and a sound playback module.

[0052] It is understandable that the first terminal and the second terminal can be two independent devices; they can be the same device or different devices. The first terminal and the second terminal can communicate via a wireless connection or a wired connection.

[0053] It is understood that a voice signal can include sound waves that propagate through a medium such as air, representing human voices. The first voice signal can refer to the voice signal collected by the first terminal. When a first user and a second user are having a close-range conversation via the first terminal and the second user is having a conversation via the second terminal, the voice signal collected by the first terminal may include the voice signal emitted by the first user, noise signals transmitted and played back by the first terminal to the second terminal and then transmitted through the air and collected by the first terminal, and noise signals transmitted and played back by the second user and then transmitted through the air and collected by the first terminal.

[0054] The second voice signal can refer to the voice signal transmitted from the first terminal to the second terminal via a wireless network and / or wired connection. Specifically, it can be the target voice signal after suppressing the first voice signal. It is understood that during communication between the first and second terminals, the respective voice signals they acquire will be continuously suppressed. Specifically, the first terminal may not perform suppression processing on the first frame of the first voice signal it acquires and will transmit it directly to the second terminal. However, subsequent frames of the first voice signal will be suppressed before being transmitted to the second terminal. Therefore, the second frame and subsequent frames of the second voice signal transmitted from the first terminal to the second terminal are all target voice signals obtained by suppressing the first frame of the first voice signal.

[0055] Specifically, in step S1, the first voice signal is acquired through the first terminal. For example... Figure 3 As shown, A and B are two users making a call. Mic1 and Speaker1 are the microphone and player of the first terminal corresponding to user A, and Mic2 and Speaker2 are the microphone and player of the second terminal corresponding to user B. The first voice signal collected by the first terminal includes the voice signal (A0) of user A speaking directly into the microphone of the first terminal, and may also include the interference voice signal A2, which may be collected by the first terminal after the first terminal transmits and plays the voice signal A0 emitted by user A to the second terminal and propagates through the air. It may also include the interference voice signal B3, which may be collected by the first terminal after the voice signal B0 emitted by user B is transmitted and played to the first terminal and propagates through the air.

[0056] In this system, the first terminal acts as user A's communication device. It is responsible for collecting user A's voice and transmitting it to the second terminal corresponding to user B. Simultaneously, it also receives user B's voice and eliminates feedback caused by acoustic feedback. The second terminal acts as user B's communication device. It is responsible for collecting user B's voice and transmitting it to user A. Simultaneously, it also receives user A's voice and eliminates feedback caused by acoustic feedback. The first voice signal refers to the raw voice signal collected by the first terminal in the full-duplex communication system, including the user's voice speaking directly into the microphone of the first terminal, as well as sound that may propagate through space from the speaker to the microphone.

[0057] The first terminal transmits the collected voice signal (including the original voice and any possible interference) to the second terminal, forming a second voice signal. This second voice signal, collected by the first terminal and transmitted to the second terminal in a full-duplex communication system, represents the voice information of the first terminal user (User A). After processing and encoding, it is sent to the second terminal via the communication network. Simultaneously, the second terminal also transmits the collected voice signal to the first terminal, forming a third voice signal. This third voice signal, collected by the second terminal and transmitted to the first terminal in the full-duplex communication system, represents the voice information of the second terminal user (User B). After processing and encoding, it is sent to the first terminal via the communication network.

[0058] S2. Based on the initial multiple adaptive filters, adaptive filtering is performed on the second and third speech signals respectively to obtain the interference signal in the simulated first speech signal; the interference signal is used to update the adaptive filters, and the updated adaptive filters are used to continue filtering the second and third speech signals in subsequent frames.

[0059] Adaptive filters, a type of filter in digital signal processing, can automatically adjust their filtering parameters based on the characteristics of the input signal to achieve optimal filtering performance. Adaptive filtering, as a signal processing technique, primarily utilizes adaptive filters to adjust their parameters, automatically optimizing filtering performance in different signal environments. In this embodiment, the adaptive filter can be adaptively updated based on the calculated interference signal to adjust its parameters.

[0060] Specifically, in step S2, after acquiring the second voice signal transmitted from the first terminal to the second terminal and the third voice signal transmitted from the second terminal to the first terminal, multiple adaptive filters are first initialized on both the first and second terminals for subsequent interference signal simulation and suppression. Then, the second and third voice signals are processed using these multiple adaptive filters to simulate interference signals that may be present in the first voice signal. These interference signals, as the target of the adaptive filtering process, need to be accurately identified and eliminated to ensure the clarity and quality of the call.

[0061] For example, an adaptive filtering algorithm can be used to simulate the generation path of interference signals. For instance, a voice signal A0 emitted by user A is transmitted to a second terminal via a wireless network and / or wired connection, and then played through the second terminal's speaker to generate voice signal A1. Voice signal A1 then propagates through a spatial propagation path and can be picked up by the microphone Mic1 of the first terminal, thereby generating voice signal A2. Similarly, a voice signal B0 emitted by user B is transmitted to a first terminal via a wireless network and / or wired connection, and then played through the first terminal's speaker to generate voice signal B1. Voice signal B1 then propagates through a spatial propagation path and can be picked up by the microphone Mic1 of the first terminal, thereby generating voice signal B3.

[0062] Therefore, the first voice signal collected by the microphone Mic1 of the first terminal includes: voice signal A0, voice signal A2, and voice signal B3 emitted by user A. Among them, voice signal A2 and voice signal B3 are interference signals compared to voice signal A0.

[0063] It is understandable that voice signal A2 is similar to the second voice signal transmitted from the first terminal to the second terminal. Therefore, the interference signal corresponding to voice signal A2 can be obtained by simulating the second voice signal. Similarly, voice signal B3 is similar to the third voice signal transmitted from the second terminal to the first terminal. Therefore, the interference signal corresponding to voice signal B3 can be obtained by simulating the third voice signal.

[0064] Specifically, an initial adaptive filter is used to simulate the second speech signal to obtain an interference signal similar to the second speech signal. Then, another initial adaptive filter is used to simulate the third speech signal to obtain an interference signal similar to the third speech signal. The interference signal is used to simulate noise generated due to acoustic feedback.

[0065] In addition, the parameters of the adaptive filter are updated based on the simulated interference signal to improve the accuracy of the simulation and prepare for filtering of the speech signal in subsequent frames.

[0066] S3. Suppress the interference signal in the first speech signal to obtain the suppressed target speech signal.

[0067] Suppression processing refers to removing interference signals from speech signals through techniques such as adaptive filters, in order to improve the clarity and quality of calls.

[0068] It is understandable that the target speech signal is the pure speech signal after filtering out interference signals from the first speech signal. As the final speech signal output by the communication system, the target speech signal directly affects the user's call experience and the quality of voice communication. After howling suppression and noise cancellation processing, the target speech signal is the pure speech that the user expects to hear, without any interference components.

[0069] Specifically, in step S3, the updated adaptive filter is used to identify and suppress interference signals in the first speech signal, thereby extracting a clear target speech signal. When the adaptive filter perfectly matches the actual feedback path, it can simulate a feedback sound that is completely consistent with the original acoustic path. At this point, the actual feedback path can be cut off, fundamentally eliminating howling.

[0070] It is understandable that speech signal processing is sequential. The first frame of the first speech signal can be considered a noise-free speech signal and is directly transmitted to the second terminal. At this time, the second frame of the second speech signal transmitted to the second terminal is almost identical to the first speech signal. The first speech signals after the second frame may contain noise interference signals. Therefore, multiple initial adaptive filters are used to adaptively filter the second and third speech signals after the second frame to obtain the interference signals in the simulated first speech signals of the corresponding frames. Then, the interference signals in the first speech signals after the second frame are suppressed to obtain the suppressed target speech signal, which is transmitted to the second terminal. The second speech signal transmitted to the second terminal is the suppressed target speech signal. Then, the processing of the first speech signal of the next frame is continuously performed based on the second and third speech signals of the previous frame.

[0071] In summary, this embodiment achieves fundamental elimination of howling by accurately simulating and suppressing feedback sounds, rather than merely reducing the loudness of howling. This allows users to enjoy a clearer and interference-free voice communication experience during calls. Furthermore, the adaptive filter can automatically adjust according to changes in the real-time call environment, maintaining a highly efficient howling suppression effect. The entire process requires no manual intervention, automatically suppressing howling and adapting to various communication environments. It is evident that this embodiment reduces instability caused by howling by eliminating howling, thereby improving the stability and reliability of the communication system.

[0072] In some embodiments, the first voice signal in this embodiment includes the voice signal emitted by the first user, the voice signal corresponding to the first interference path between the first terminal and the second terminal, and the voice signal corresponding to the second interference path of the first terminal itself.

[0073] Interference paths typically refer to various unwanted signal paths encountered by a speech signal during transmission or recording. These signals degrade the quality of the speech signal, affecting its clarity and intelligibility. Interference paths can include the following types: echoes, reverberation, ambient noise, electronic equipment noise, interfering sound sources, radio frequency interference, etc.

[0074] The first interference path can be: sound played by user A through the speaker of the first terminal propagates through space to the microphone of the first terminal, forming a path A1 to A2; and sound played by user B through the speaker of the second terminal propagates through space to the microphone of the second terminal, forming a path B1 to B2, such as... Figure 3 Two paths in the middle.

[0075] The second interference path could be: sound played by user A through the speaker of the first terminal propagates through space and is picked up by the microphone of the first terminal, forming a path from B1 to B3; and sound played by user B through the speaker of the second terminal propagates through space and is picked up by the microphone of the second terminal, forming a path from A1 to A3, such as... Figure 3 Path 3 and Path 3' in the text.

[0076] Specifically, in this embodiment, the first voice signal obtained by the first terminal is mainly the sound of the user speaking directly into the microphone of the first terminal. It may also include the voice signal (A2) corresponding to the first interference path between the first terminal and the second terminal, and the voice signal (B3) corresponding to the second interference path of the first terminal itself. Here, A2 is the sound played by the speaker of the second terminal, which is the signal collected by the microphone of the first terminal after propagating through space; B3 is the sound played by the speaker of the first terminal, which is also the signal collected by the microphone of the first terminal after propagating through space.

[0077] This embodiment can effectively reduce or eliminate howling and echo by accurately identifying and suppressing interference signals in the first speech signal, thereby significantly improving the clarity and quality of the call.

[0078] In some embodiments, the first interference path in this embodiment is the path for transmitting voice signals through space between the first terminal and the second terminal; the second interference path is the path for transmitting voice signals through space between the first terminal and the second terminal.

[0079] The path for transmitting voice signals through space can be: user A plays sound through the speaker of the first terminal, and the sound travels through the air in the space and is captured by the microphone of the first terminal; or user B plays sound through the speaker of the second terminal, and the sound travels through the air in the space where user A is located and is captured by the microphone of the first terminal corresponding to user A.

[0080] Specifically, in this embodiment, the first interference path can refer to the path of transmitting voice signals between the first terminal and the second terminal through space, that is, the path of the sound played by the far-end (second terminal) speaker (Speaker2) propagating through the air to the near-end (first terminal) microphone (Mic1).

[0081] The second interference path can refer to the path through which the first terminal transmits voice signals in space, that is, the path through which the sound played by the near-end (first terminal) speaker (Speaker1) travels through the air to its own microphone (Mic1).

[0082] This embodiment, by clearly distinguishing between two interference paths, can more accurately identify and differentiate the source of interference signals, thereby enabling more effective suppression. Since howling and echo may originate from different paths, clearly identifying these interference paths helps the adaptive filter to more accurately simulate and cancel these interferences, improving the howling suppression effect.

[0083] In some embodiments, the second voice signal includes a voice signal transmitted from the first terminal to the second terminal via a target path; the third voice signal includes a voice signal transmitted from the second terminal to the first terminal via a target path.

[0084] Specifically, in this embodiment, the second voice signal can be the voice signal transmitted from the first terminal to the second terminal through the target path, and the third voice signal can be the voice signal transmitted from the second terminal to the first terminal through the target path. The target path can be considered as a clean, interference-free voice signal path, which can be the path through which the first terminal and the second terminal transmit signals via a wired / wireless network.

[0085] This embodiment distinguishes between the second and third voice signals, clarifying the direction of signal flow between the two terminals. This helps to better understand and control the signal transmission process, thereby enabling the adoption of different processing strategies for signals in different directions and improving the effectiveness of voice signal processing.

[0086] In some embodiments, the adaptive filter includes a first adaptive filter and a second adaptive filter; the interference signal includes a first interference signal and a second interference signal;

[0087] Step S2, "Based on the initial multiple adaptive filters, adaptive filtering is performed on the second and third speech signals respectively to obtain the interference signal in the simulated first speech signal," may specifically include:

[0088] S21. The second speech signal is filtered by the first adaptive filter to obtain the first interference signal in the simulated first speech signal that corresponds to the first interference path;

[0089] S22. The third speech signal is filtered by the second adaptive filter to obtain the second interference signal in the simulated first speech signal that corresponds to the second interference path.

[0090] It is understood that the adaptive filter in this embodiment may include a first adaptive filter and a second adaptive filter. The first adaptive filter is used to simulate the second speech signal to simulate the first interference signal in the first speech signal; the second adaptive filter is used to simulate the third speech signal to simulate the second interference signal in the first speech signal.

[0091] The first interference signal can be: after the first terminal transmits the sound signal A0 emitted by user A to the second terminal and plays the emitted sound signal A1, the sound signal A1 propagates through the air (e.g., Figure 3 The interference voice signal A2 is collected by the microphone of the first terminal via path 2). Therefore, the first interference signal in the first voice signal collected by the first terminal can be the interference voice signal A2.

[0092] It is understandable that the first terminal will suppress the first voice signal it collects, and the second terminal will also suppress the voice signal it collects. Similarly, for the second terminal, the voice signal it collects may also include two types of interference signals. The first interference signal could be: after the second terminal transmits the voice signal B0 emitted by user B to the first terminal and plays the emitted voice signal B1, the voice signal B1 propagates through the air (e.g., ...). Figure 3 The interference voice signal B2 is collected by the microphone of the second terminal via path 2).

[0093] The second interference signal could be: after the second terminal transmits the sound signal B0 emitted by user B to the first terminal and plays the emitted sound signal B1, the sound signal B1 propagates through the air (e.g., Figure 3 The interference voice signal B3 is collected by the microphone of the first terminal in path 3). Therefore, the second interference signal in the first voice signal collected by the first terminal can be the interference voice signal B3.

[0094] Similarly, for the second terminal, the second interference signal in the voice signal collected by the second terminal could be: after the first terminal transmits the voice signal A0 emitted by user A to the second terminal and plays the emitted voice signal A1, the voice signal A1 propagates through the air (e.g., Figure 3 The interference voice signal A3 is collected by the microphone of the second terminal via path 3').

[0095] The first adaptive filter is used to eliminate or reduce interference signals (i.e., interference speech signal A2) in the speech signal acquired by the first terminal, caused by the sound emitted by the speaker of the second terminal propagating through a spatial path and being acquired by the microphone of the first terminal. The second adaptive filter is used to eliminate or reduce interference signals (i.e., interference speech signal B3) in the speech signal acquired by the first terminal, caused by the sound emitted by the speaker of the first terminal propagating through a spatial path and being acquired by the microphone of the same first terminal.

[0096] Specifically, the adaptive filter in this embodiment may include a first adaptive filter of the first terminal and a second adaptive filter of the second terminal. First, the corresponding adaptive filters are initialized on the first terminal and the second terminal respectively to simulate and suppress interference signals. The first adaptive filter is responsible for simulating and suppressing interference signals from the first interference path, and the second adaptive filter is responsible for simulating and suppressing interference signals from the second interference path. When the first terminal receives the second voice signal, it uses the first adaptive filter to filter it to simulate the first interference signal. When the first terminal receives the third voice signal, it uses the second adaptive filter to filter it to simulate the second interference signal.

[0097] In this embodiment, each adaptive filter simulates the interference signal corresponding to the corresponding interference path based on its parameters. The interference signal refers to the noise component that appears in the original speech signal. By accurately simulating and suppressing the interference signal, the clarity and quality of the call are significantly improved. Subsequently, the parameters of each adaptive filter are updated based on the simulated interference signal to improve the accuracy of the simulation and the filtering effect.

[0098] In some embodiments, step S21, "filtering the second speech signal using a first adaptive filter to obtain a first interference signal corresponding to the first interference path in the simulated first speech signal," includes:

[0099] The second speech signal is filtered by the first adaptive filter according to the current filter parameters to obtain the first interference signal in the simulated first speech signal that corresponds to the first interference path; the first interference signal is an analog signal with similar characteristics to the second speech signal.

[0100] In an adaptive filter, the filter parameters refer to a series of values ​​that define and adjust the filter's behavior, such as filter coefficients, step size parameters, and error feedback coefficients. These parameters determine how the filter responds to signals with different frequency components. In an adaptive filter, these parameters can be dynamically adjusted according to the characteristics of the input signal to optimize the filtering effect.

[0101] Among them, the analog signal with similar characteristics refers to the analog signal that has similar frequency components as the actual interference signal, similar time waveform as the actual interference signal, and similar amplitude variation as the actual interference signal.

[0102] Specifically, in this embodiment, the second speech signal is filtered by the first adaptive filter according to the current filter parameters. The simulated first interference signal obtained by the filtering process is a signal with similar characteristics to the second speech signal. The first interference signal represents the interference transmitted back to the first terminal microphone through the first interference path (i.e., the spatial path).

[0103] In the specific implementation process, the parameters of the first adaptive filter are initialized by the first terminal so that the filter adapts to the expected characteristics of the interference signal; the first terminal receives the second voice signal from the second terminal, the second voice signal being the signal transmitted through the target path; the first adaptive filter is used to filter the second voice signal to simulate the first interference signal, the first interference signal simulating the echo or noise that may be generated when the second voice signal propagates through the first interference path.

[0104] This embodiment ensures that the first adaptive filter can accurately simulate the interference signal generated by the first interference path. The accurately simulated interference signal can be more effectively suppressed, thereby significantly reducing howling and improving call quality. This makes it so that users can hardly feel the existence of interference signals during the call, effectively improving the call experience.

[0105] In some embodiments, step S22, "filtering the third speech signal using a second adaptive filter to obtain a second interference signal corresponding to the second interference path in the simulated first speech signal," includes:

[0106] The third speech signal is filtered by the second adaptive filter according to the current filter parameters to obtain the second interference signal in the simulated first speech signal that corresponds to the second interference path; the second interference signal is a simulated signal with similar characteristics to the third speech signal.

[0107] Filter parameters refer to a series of values ​​that define and adjust the behavior of the filter, such as filter coefficients, step size parameters, and error feedback coefficients. These parameters determine how the filter responds to signals with different frequency components. In adaptive filters, these parameters can be dynamically adjusted according to the characteristics of the input signal to optimize the filtering effect.

[0108] Specifically, in this embodiment, the third speech signal is filtered by the second adaptive filter according to the current filter parameters. The simulated second interference signal obtained by the filtering process is a signal with similar characteristics to the third speech signal. The second interference signal represents the interference transmitted back to the microphone of the first terminal through the second interference path (i.e., the spatial path inside the first terminal).

[0109] In a specific embodiment, the parameters of the second adaptive filter are initialized by the first terminal so that the filter adapts to the expected characteristics of the internal interference signal; the first terminal receives a third voice signal from the second terminal, the third voice signal being a signal transmitted through a target path (which may be wired or wireless); the third voice signal is filtered using the second adaptive filter to simulate a second interference signal, the second interference signal being used to simulate the echo or noise that may be generated when the third voice signal propagates through the spatial path inside the first terminal.

[0110] This embodiment ensures that the second adaptive filter can accurately simulate the internal interference signal generated by the second interference path, so that the accurately simulated interference signal can be suppressed more effectively, thereby significantly reducing howling and echo, improving call quality. Therefore, users can hardly feel the presence of the internal interference signal during the call, effectively improving the user's call experience.

[0111] In some embodiments, the speech signal processing method further includes:

[0112] Based on the first interference signal and the voice signal corresponding to the first interference path between the first terminal and the second terminal, update the first adaptive filter corresponding to the first interference path.

[0113] Based on the second interference signal and the voice signal corresponding to the second interference path of the first terminal itself, the second adaptive filter corresponding to the second interference path is updated.

[0114] Specifically, the speech signal processing method in this embodiment also provides a method for updating each adaptive filter based on interference signals. First, during a call, a simulated first interference signal is obtained by processing the second speech signal through a first adaptive filter, and a simulated second interference signal is obtained by processing the third speech signal through a second adaptive filter. By analyzing the characteristics corresponding to the first and second interference signals, including frequency, amplitude, and time delay, the adjustment direction of the parameters of the first and second adaptive filters is determined. Based on the analysis results of the interference signals, the parameters of each adaptive filter are adjusted to improve the performance of subsequent simulation and suppression of interference signals.

[0115] In addition, a real-time update mechanism can be designed to ensure that the filter parameters can be dynamically adjusted according to real-time feedback during the call, and to evaluate the filter performance to ensure that the updated filter can effectively simulate and suppress interference signals.

[0116] This embodiment improves the adaptability and howling suppression effect of the filter by ensuring that the adaptive filter can be adjusted according to real-time call conditions. By accurately updating the filter parameters, call quality is significantly improved and howling and echo are reduced. At the same time, dynamically adjusting the filter parameters helps to optimize the overall performance of the call system and improve the efficiency of signal processing.

[0117] In some embodiments, the interference signal includes a first interference signal and a second interference signal; the interference signal in the first speech signal is suppressed to obtain a suppressed target speech signal, including:

[0118] The first interference signal and the second interference signal in the first speech signal are suppressed to remove the first interference signal and the second interference signal from the first speech signal, so as to obtain the suppressed target speech signal.

[0119] Specifically, after filtering the second speech signal using the first adaptive filter to obtain the first interference signal corresponding to the first interference path in the simulated first speech signal, and filtering the third speech signal using the second adaptive filter to obtain the second interference signal corresponding to the second interference path in the simulated first speech signal, it is necessary to suppress the first and second interference signals in the first speech signal and suppress the extracted interference signals. After subtracting the first and second interference signals from the first speech signal, the target speech signal, i.e., the clean speech that the user wants to transmit, is reconstructed.

[0120] In addition, this embodiment can also update the parameters of the first adaptive filter and the second adaptive filter based on the result of the suppression processing, so as to more accurately simulate and suppress interference signals in future call frames.

[0121] This embodiment significantly improves call clarity and quality by effectively suppressing interference signals. The adaptive filter can adjust according to real-time call conditions, improving the filter's performance and thus adapting to various communication scenarios, including close-range calls, long-range calls, and different communication environments.

[0122] To better understand the speech signal processing method provided in this embodiment, this embodiment also provides a specific implementation method for the speech signal processing method.

[0123] like Figure 3 As shown, A and B are two users making a call. Mic1 and Speaker1 are the microphone and player of the first terminal corresponding to user A, and Mic2 and Speaker2 are the microphone and player of the second terminal corresponding to user B.

[0124] When the first terminal and the second terminal are communicating, user A speaks to Mic1, and the voice signal collected by Mic1 is recorded as A0. The voice signal A0 travels through the wired or wireless network path between the first terminal and the second terminal (e.g., ...). Figure 3 The audio signal A1 is transmitted via path 1) to Speaker2 of the second terminal, and then broadcast by Speaker2. The audio signal A1 broadcast by Speaker2 travels through the spatial path between user A and user B (e.g., path 1) to become speech signal A1. Figure 3 Path 2) can be acquired by the Mic1 of the first terminal, therefore the first speech signal acquired by the Mic1 of the first terminal includes speech signal A2.

[0125] User B speaks into Mic2 on the second terminal, and the voice signal captured by Mic2 is recorded as B0. The voice signal B0 travels through a wired or wireless network path between the first and second terminals (e.g., ...). Figure 3 The signal is transmitted via path 1) to Speaker1 of the first terminal. After being played by Speaker1, it becomes voice signal B1. Voice B1 travels through the spatial path between the first terminals (e.g., ...). Figure 3 The first speech signal collected by the first terminal's Mic1 is then collected via path 3), therefore the first speech signal collected by the first terminal's Mic1 includes speech signal B3.

[0126] In this scenario, the sound collected by the Mic1 of the first terminal includes voice signal A0 (the voice that user A wants to send out) + voice signal A2 (the voice that A1 transmits through the Speaker2 of the second terminal and then propagates to the Mic1 of the first terminal via path 2) + voice signal B3 (the voice B1 emitted by user B, which is then transmitted through the Speaker1 of the first terminal and collected by the Mic1 of the first terminal). Among these, voice signals A2 and B3 are interference signals.

[0127] During a continuous voice call between users A and B, the audio is constantly amplified and oscillated between the microphone and speaker of the first terminal and the microphone and speaker of the second terminal, easily causing feedback during the call. If interference signals A2 and B3 are removed from the first audio signal collected by the microphone 1 of the first terminal, a noise-free audio signal A0 can be obtained. The first terminal then transmits the interference-suppressed audio signal A0 to the second terminal via a wired or wireless network path. The audio signal A1 played through the speaker 2 of the second terminal is the cleaner audio signal that user A wants user B to hear, corresponding to audio signal A0.

[0128] Similarly, when user A and user B are talking, user B speaks into the microphone 2 of the second terminal. The voice captured by the microphone 2 is recorded as B0. B0 is transmitted to the speaker 1 of the first terminal via path 1. After being played back by the speaker 1 of the first terminal, it becomes B1. The voice B1 played back by the speaker 1 of the first terminal is transmitted through path 2 and can be captured by the microphone 2 of the second terminal. At this time, the microphone 2 of the second terminal captures the voice B2 generated based on the voice B1.

[0129] User A speaks into the microphone Mic1 of the first terminal, and the voice signal collected by Mic1 of the first terminal is recorded as A0. The voice signal A0 passes through a circuit (such as...). Figure 3 Path 1) is transmitted to Speaker 2 of the second terminal of user B, and becomes voice signal A1; after voice signal A1 is played through Speaker 2 of the second terminal, it travels through the spatial path between Speaker 2 and Mic 2 of the second terminal (e.g., ... Figure 3 After propagation through path 3'), the speech signal is collected by the second terminal's Mic2, and the collected speech signal is speech signal A3.

[0130] The voice signal collected by Mic2 on User B's second terminal may include: voice signal B0 (the voice that User B wants to send out) + voice signal B2 (the voice that B1 transmits through the space between Users A and B to Mic2 on the second terminal after passing through Speaker1 on the first terminal) + voice signal A3 (the voice A1 sent by User A is collected by Mic2 on the second terminal after passing through Speaker2 on User B's second terminal). Among them, voice signals B2 and A3 are interference voice signals.

[0131] Similarly, the situation is the same for terminal B (the second terminal) and terminal A (the first terminal). The second terminal also removes interference voice signals B2 and A3 from the voice signal collected by its Mic2, thus obtaining voice signal B0, which is the voice B1 that user B wants user A to hear. The second terminal transmits voice signal B0 to the first terminal through a wired or wireless network path between the first and second terminals. The voice signal B1 broadcast by the first terminal's Speaker1 is the cleaner voice signal corresponding to voice signal B0 that user B wants user A to hear.

[0132] The speech signal processing method provided in this embodiment uses the idea of ​​feedback cancellation to control the acoustic path feedback. This mainly involves subtracting the feedback signal component from the microphone signal to obtain the residual E(ω) / E'(ω), which is the speech signal after removing interference signals. The specific process is as follows:

[0133] like Figures 4-5 As shown, during a call between terminals A and B, taking terminal A as the executing entity, terminal A's microphone picks up the voice signal "A", while terminal B's microphone picks up the voice signal "B". Terminal A will transmit the interference-suppressed signal "A1" to terminal B, and terminal A will also receive the interference-suppressed voice signal "B1" transmitted by terminal B.

[0134] The voice signal A collected by terminal A is the first voice signal, including: A0+A2+B3. Here, voice signal A0 [represents the voice that user A wants to send out (transmitted through path 1 between the first and second terminals)], voice signal A2 [voice signal A0 is transmitted to the second terminal and then played out as voice signal A1 by the second terminal; voice signal A1 passes through Speaker2 of the second terminal and is then transmitted through the spatial path between the first and second terminals (e.g., ...]]. Figure 3 The voice signal A2, which is collected by the Mic1 of the first terminal, and the voice signal B3, which is transmitted from user B to the first terminal and then played out by the first terminal, are collected by the Mic1 of the first terminal after being played by the Speaker1 of the first terminal.

[0135] Simultaneously, end A will receive the interference-suppressed voice signal B1 (the second voice signal) transmitted by end B via path 1. (Path 1 can be a wired circuit or a stable wireless network link.)

[0136] Specifically, at end A, the speech signal "A1" can be input to the first adaptive filter (filter1(ω)) for processing. The adaptive rate filter filters the input speech signal "A1" according to the current filter parameters. Specifically, the speech signal A1 can be simulated and estimated to obtain the filtered output signal (first estimated signal). The output signal is the interference speech signal A2 in the simulated signal A, which is similar to A1 and is transmitted to end A Mic1 through the spatial path (path 2) between B and Speaker2.

[0137] Meanwhile, the received speech signal "B1" is input to the second adaptive filter (filter2(ω)) for processing. The adaptive rate filter filters the input speech signal "B1" according to the current filter parameters, that is, it performs simulated estimation on the speech signal B1 to obtain the filtered output signal (second estimated signal). The output signal is the interference speech signal in the simulated signal A that is similar to the speech signal B0, that is, the interference speech signal B3 in the first speech signal A.

[0138] Then, the interfering speech signals A2 and B3 in the first speech signal "A" are subtracted (i.e., suppressed) to obtain the target speech signal after interference suppression.

[0139] Similarly, for terminal B, i.e. the second terminal, the second terminal inputs the voice signal B1 to the first adaptive filter (filter1(ω)) for adaptive filtering. The adaptive rate filter filters the input voice signal "B1" according to the current filter parameters. Specifically, the filtered output signal can be obtained by simulating and estimating the voice signal B1. The output signal is the interference voice signal B2 in the voice signal B, which is similar to B1 and is transmitted to terminal B Mic2 through the spatial path (path 2) between A and B after passing through Speaker1 at terminal A.

[0140] Meanwhile, the received speech signal "A1" is input to the second adaptive filter (filter2'(ω)) for processing. The adaptive rate filter filters the input speech signal "B1" according to the current filter parameters, that is, it performs simulation estimation on the speech signal A1 to obtain the filtered output signal. The output signal is the interference speech signal in the simulated signal B that is similar to the speech signal A1, that is, the interference speech signal A3 in the speech signal B.

[0141] Since both the first adaptive filter used by the first terminal and the first adaptive filter used by the second terminal simulate the interference speech signal corresponding to the spatial propagation path between the first and second terminals, the first adaptive filter used by the first terminal and the first adaptive filter used by the second terminal can be the same, i.e., both can use filter1(ω). However, the second adaptive filter used by the first terminal simulates the interference speech signal corresponding to the spatial propagation path between the speaker and microphone of the first terminal (i.e., the spatial propagation path around the first terminal itself), while the second adaptive filter used by the second terminal simulates the interference speech signal corresponding to the spatial propagation path between the speaker and microphone of the second terminal (i.e., the spatial propagation path around the second terminal itself). Therefore, the second adaptive filter filter2(ω) used by the first terminal and the second adaptive filter filter2'(ω) used by the second terminal are different.

[0142] After the first terminal suppresses the acquired first speech signal, it updates the adaptive filters filter1(ω) and filter2(ω). It then estimates the actual feedback paths G1(ω) and G2(ω) based on the input speech signal of the next frame. When filter1(ω) and G1(ω) match perfectly, and G2(ω) and filter2(ω) match perfectly, the resulting re-feedback sounds filter_A2 and A2, and filter_B3 and B3 are completely identical. This indicates that passing through the filters is equivalent to following the original acoustic path, thus achieving acoustic path estimation. At this point, the feedback path can be cut off, fundamentally eliminating the howling problem caused by the feedback sound. The feedback paths G1(ω) and G2(ω) can be estimated through the filters.

[0143] After the second terminal suppresses the acquired speech signal, it updates the adaptive filters filter1(ω) and filter2'(ω). It then estimates the actual feedback paths G1(ω) and G2'(ω) based on the next frame of the input speech signal. When filter1(ω) and G1(ω) match perfectly, and G2'(ω) and filter2'(ω) match perfectly, the resulting re-feedback sounds filter_B2 and B2, and filter_A3 and A3 are completely identical. This indicates that passing through the filters is equivalent to following the original acoustic path, thus achieving acoustic path estimation. At this point, the feedback path can be cut off, fundamentally eliminating the howling problem caused by the feedback sound. The feedback paths G1(ω) and G2'(ω) can be estimated through the filters.

[0144] The filter update process is as follows: an evaluation criterion for adaptive filters, such as standard deviation, is used. Based on the evaluation results, the filter parameters are adjusted to achieve adaptive filter updates, so that the speech signal estimated by the adaptive filter is closer to the interference signal.

[0145] like Figure 6 As shown, filter1(ω), filter2(ω), and filter2'(ω) are the adaptive filters added to the entire system in this scheme to suppress howling. Specifically, filter1(ω) [the filter corresponding to path 2 at end A], filter2(ω) [the filter corresponding to path 3 at end A], and filter2'(ω) [the filter corresponding to path 3 at end B]. The filter corresponding to path 2 at end B can be the same as filter1(ω) at end A.

[0146] For example, in a practical application scenario, taking the communication between the screen inside the door and the peephole outside the door as an example, the following diagram illustrates the situation. Figure 6 As shown. This application scenario applies to calls between two devices connected via a wired connection, and also to devices connected wirelessly.

[0147] Taking either the inside or outside of the door as an example, the process of generating a whistling sound is as follows: Figure 6 As shown, G1(ω) represents Figure 3 The transfer function corresponding to path 2 in the code. Figure 5 The adaptive filter fliter1(ω) in the image is the reconstructed image. Figure 3 If the transfer function G1(ω) of path 2 is infinitely close to the adaptive filter filter1(ω) through its adaptivity, then the speech filterA2(ω) after passing through filter1(ω) is infinitely close to the speech A2(ω) after passing through path 2.

[0148] Similarly, Figure 5 In the adaptive filter2(ω) reconstruction diagram, the transfer function G2(ω) of path 3 is infinitely close to the transfer function G2(ω) through the adaptive filter filter2(ω). Therefore, the speech filterB3(ω) after passing through filter2(ω) is infinitely close to the speech B3(ω) after passing through path 3.

[0149] Without an adaptive filter, the first speech signal A(ω) acquired by the microphone at end A includes:

[0150] A(ω)=A0(ω)+A2(ω)+B3(ω),

[0151] The first voice signal A(ω) is transmitted to terminal B via path 1 and played through the speaker at terminal B. Amplification may cause a feedback sound. Here, A0(ω) represents the voice signal A0 emitted by user A, collected by the first terminal. A2(ω) represents the voice signal A1 generated by the second terminal after the voice signal A0 is transmitted via a wireless network and / or wired connection, and then played through the second terminal's speaker. The voice signal A1 then propagates through a spatial propagation path and can be collected by the microphone Mic1 of the first terminal, resulting in the interference voice signal A2. B3(ω) represents the voice signal B3 generated by the first terminal after the voice signal B0 emitted by user B is transmitted via a wireless network and / or wired connection, and then played through the first terminal's speaker. The voice signal B1 then propagates through a spatial propagation path and can be collected by the microphone Mic1 of the first terminal.

[0152] After adding adaptive filters filter1(ω) and filter2(ω), the suppressed target speech signal E(ω) is obtained by suppressing the first speech signal A(ω):

[0153] E(ω)=A(ω)-filterA2(ω)-filterB3(ω),

[0154] That is, E(ω)=A0(ω)+A2(ω)+B3(ω)-filterA2(ω)-filterB3(ω).

[0155] At this point, E(ω)≈A0(ω), achieving the goal of transmitting a relatively pure voice signal.

[0156] Where filterA2(ω) represents the output of signal A1(ω) after suppression by the adaptive filter filter1(ω);

[0157] filterB3(ω): represents the output of signal B1(ω) after being suppressed by the adaptive filter filter2(ω).

[0158] Similarly, the adaptive filter of the second terminal uses the same working principle, which will not be elaborated here.

[0159] In summary, the speech signal processing method provided in this application adds multiple adaptive filters to a full-duplex communication system and adopts a feedback cancellation approach. During a full-duplex call, the adaptive filters at both the near and far ends are simultaneously invoked to predict the actual feedback path and match it with the adaptive filters, thereby cutting off the feedback path and suppressing howling during the call. This effectively optimizes the short-distance call effect, avoids the significant reverberation tail caused by howling, and thus effectively improves the quality of the call voice.

[0160] To facilitate better implementation of the speech signal processing method of this application, this application also provides a speech signal processing apparatus based on the above-described speech signal processing method. The meanings of the terms used are the same as in the speech signal processing method described above, and specific implementation details can be found in the descriptions within the method embodiments.

[0161] Please see Figure 7 , Figure 7 This is a schematic diagram of the structure of a speech signal processing device provided in an embodiment of this application. The speech signal processing device includes an acquisition module 201, a filtering module 202, and a suppression module 203, which can be specifically described as follows:

[0162] The acquisition module 201 is used to acquire a first voice signal collected by the first terminal, a second voice signal transmitted from the first terminal to the second terminal, and a third voice signal transmitted from the second terminal to the first terminal;

[0163] The filtering module 202 is used to perform adaptive filtering on the second and third speech signals based on the initial multiple adaptive filters to obtain the interference signal in the simulated first speech signal; the interference signal is used to update the adaptive filters, and the updated adaptive filters are used to continue filtering the second and third speech signals in subsequent frames;

[0164] Suppression module 203 is used to suppress interference signals in the first speech signal to obtain the suppressed target speech signal;

[0165] In some embodiments, the first voice signal includes a voice signal emitted by the first user, a voice signal corresponding to a first interference path between the first terminal and the second terminal, and a voice signal corresponding to a second interference path of the first terminal itself.

[0166] In some embodiments, the first interference path is the path for transmitting voice signals through space between the first terminal and the second terminal; the second interference path is the path for transmitting voice signals through space between the first terminal and the second terminal.

[0167] In some embodiments, the second voice signal includes a voice signal transmitted from the first terminal to the second terminal via a target path; the third voice signal includes a voice signal transmitted from the second terminal to the first terminal via a target path.

[0168] In some embodiments, the adaptive filter includes a first adaptive filter and a second adaptive filter; the interference signal includes a first interference signal and a second interference signal;

[0169] The filtering module includes:

[0170] The first filtering unit is used to filter the second speech signal through a first adaptive filter to obtain the first interference signal in the simulated first speech signal corresponding to the first interference path.

[0171] The second filtering unit is used to filter the third speech signal through the second adaptive filter to obtain the second interference signal in the simulated first speech signal that corresponds to the second interference path.

[0172] In some embodiments, the first filtering unit is specifically used to filter the second speech signal according to the current filter parameters using a first adaptive filter to obtain a first interference signal in the simulated first speech signal that corresponds to the first interference path; the first interference signal is an analog signal that has similar characteristics to the second speech signal.

[0173] In some embodiments, the second filtering unit is specifically used to filter the third speech signal according to the current filter parameters using a second adaptive filter to obtain a second interference signal in the simulated first speech signal that corresponds to the second interference path; the second interference signal is an analog signal that has similar characteristics to the third speech signal.

[0174] In some embodiments, the apparatus further includes an update module, the update module comprising:

[0175] The first update unit is used to update the first adaptive filter corresponding to the first interference path based on the first interference signal and the voice signal corresponding to the first interference path between the first terminal and the second terminal.

[0176] The second update unit is used to update the second adaptive filter corresponding to the second interference path based on the second interference signal and the voice signal corresponding to the second interference path of the first terminal itself.

[0177] In some embodiments, the interference signal includes a first interference signal and a second interference signal; the suppression module is specifically used to: perform suppression processing on the first interference signal and the second interference signal in the first speech signal to remove the first interference signal and the second interference signal from the first speech signal to obtain the suppressed target speech signal.

[0178] As described above, this application provides a speech signal processing device. An acquisition module acquires a first speech signal collected by a first terminal, a second speech signal transmitted from the first terminal to a second terminal, and a third speech signal transmitted from the second terminal to the first terminal. A filtering module performs adaptive filtering on the second and third speech signals based on initial multiple adaptive filters to obtain interference signals in the simulated first speech signal. The interference signals are used to update the adaptive filters, and the updated adaptive filters are used to continue filtering the second and third speech signals in subsequent frames. A suppression module suppresses the interference signals in the first speech signal to obtain a suppressed target speech signal. This embodiment adds multiple adaptive filters to a full-duplex call system, employing a feedback cancellation approach. During a full-duplex call, adaptive filters at both near and far ends are simultaneously invoked to predict the actual feedback path and match it with the adaptive filters, thereby cutting off the feedback path and suppressing howling during the call. This effectively optimizes short-distance call performance, avoids significant reverberation tails caused by howling, and thus effectively improves call voice quality.

[0179] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 8 As shown, the computer device includes a processor, memory, network interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a voice signal processing method. The display screen can be a liquid crystal display (LCD) or an electronic ink display (EIM). The input device can be a touch layer covering the display screen or a computer device itself.

[0180] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0181] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0182] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0183] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0184] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0185] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0186] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A speech signal processing method, characterized in that, include: Acquire a first voice signal collected by a first terminal; wherein the first voice signal includes a voice signal emitted by a first user, an interference signal formed by a voice signal played by a second terminal through a first interference path between the first terminal and the second terminal, and an interference signal formed by a voice signal played by the first terminal through a second interference path of the first terminal itself. Acquire the second voice signal transmitted from the first terminal to the second terminal, and the third voice signal transmitted from the second terminal to the first terminal; The second speech signal is filtered by a first adaptive filter to obtain a first interference signal in the simulated first speech signal that corresponds to the first interference path; the third speech signal is filtered by a second adaptive filter to obtain a second interference signal in the simulated first speech signal that corresponds to the second interference path; the interference signal is used to update the adaptive filter, and the updated adaptive filter is used to continue filtering the second and third speech signals in subsequent frames; The first interference signal and the second interference signal in the first speech signal are suppressed to obtain the suppressed target speech signal.

2. The speech signal processing method as described in claim 1, characterized in that, The first interference path is the path through which the first terminal and the second terminal transmit voice signals via space; the second interference path is the path through which the first terminal and the first terminal transmit voice signals via space.

3. The speech signal processing method as described in claim 1, characterized in that, The second voice signal includes the voice signal transmitted from the first terminal to the second terminal via the target path; the third voice signal includes the voice signal transmitted from the second terminal to the first terminal via the target path.

4. The speech signal processing method as described in claim 1, characterized in that, The step of filtering the second speech signal using the first adaptive filter to obtain the first interference signal corresponding to the first interference path in the simulated first speech signal includes: The first adaptive filter filters the second speech signal according to the current filter parameters to obtain a first interference signal in the simulated first speech signal that corresponds to the first interference path; the first interference signal is a simulated signal with similar characteristics to the second speech signal.

5. The speech signal processing method as described in claim 1, characterized in that, The step of filtering the third speech signal using the second adaptive filter to obtain the second interference signal corresponding to the second interference path in the simulated first speech signal includes: The second adaptive filter filters the third speech signal according to the current filter parameters to obtain a second interference signal in the simulated first speech signal that corresponds to the second interference path; the second interference signal is an analog signal with similar characteristics to the third speech signal.

6. The speech signal processing method as described in claim 1, characterized in that, The method further includes: Based on the first interference signal and the voice signal corresponding to the first interference path between the first terminal and the second terminal, update the first adaptive filter corresponding to the first interference path. Based on the second interference signal and the voice signal corresponding to the second interference path of the first terminal itself, the second adaptive filter corresponding to the second interference path is updated.

7. The speech signal processing method as described in claim 1, characterized in that, The interference signal includes a first interference signal and a second interference signal; the suppression processing of the first interference signal and the second interference signal in the first speech signal to obtain the suppressed target speech signal includes: The first interference signal and the second interference signal in the first speech signal are suppressed to remove the first interference signal and the second interference signal from the first speech signal, so as to obtain the suppressed target speech signal.

8. A speech signal processing device, characterized in that, include: The acquisition module is used to acquire a first voice signal collected by a first terminal, a second voice signal transmitted from the first terminal to a second terminal, and a third voice signal transmitted from the second terminal to the first terminal; wherein, the first voice signal includes a voice signal emitted by a first user, an interference signal formed by the voice signal played by the second terminal through a first interference path between the first terminal and the second terminal, and an interference signal formed by the voice signal played by the first terminal through a second interference path of the first terminal itself. A filtering module is used to filter the second speech signal using a first adaptive filter to obtain a first interference signal in the simulated first speech signal that corresponds to the first interference path; and to filter the third speech signal using a second adaptive filter to obtain a second interference signal in the simulated first speech signal that corresponds to the second interference path; the interference signal is used to update the adaptive filter, and the updated adaptive filter is used to continue filtering the second and third speech signals in subsequent frames. The suppression module is used to suppress the first interference signal and the second interference signal in the first speech signal to obtain the suppressed target speech signal.

9. An electronic device comprising at least one processor and at least one memory, wherein, The memory stores a computer program, characterized in that when the processor executes the computer program, it implements the steps of the speech signal processing method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the speech signal processing method according to any one of claims 1-7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the speech signal processing method according to any one of claims 1 to 7.

Images