Far-field pickup and interference suppression system based on structure and acoustics collaborative optimization

Abstract

The invention discloses a far-field pickup and interference suppression system based on structure and acoustics collaborative optimization, and the system comprises a sound guide structure assembly, a noise isolation assembly, a microphone array, an acoustics processing module, a system control unit, and a structure-acoustics collaborative module. The sound guide structure assembly is integrated on the equipment body and is used for guiding sound waves in a target direction to the microphone array; the noise isolation assembly is arranged between the microphone array and the equipment shell and is used for attenuating structure propagation vibration; and the acoustic processing module is used for carrying out beam forming, noise suppression and echo cancellation processing on audio signals collected by the microphone array. The beneficial effects of the invention are that integrated design is carried out from a physical structure to a processing algorithm, and structural parameters are introduced to an acoustic processing link; dynamic adaptation and joint optimization of hardware characteristics and a software algorithm are realized, so that the quality and robustness of far-field speech capture are fundamentally improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent conference terminal technology, specifically to a far-field sound pickup and interference suppression system based on structural and acoustic co-optimization. Background Technology

[0002] With the normalization of remote collaborative work and online meetings, all-in-one conference machines that integrate display, interaction, and audio / video capture have become core equipment in modern conference rooms. Among them, the quality of audio capture directly determines the auditory experience and communication efficiency of remote participants. However, in the complex real office environment, the sound pickup system of existing all-in-one conference machines faces many technical bottlenecks and challenges.

[0003] Regarding pickup distance, due to limitations in device size and microphone sensitivity, the effective pickup distance of existing devices is typically between 3 and 5 meters. Beyond this range, the voice signal attenuates severely, and the signal-to-noise ratio drops sharply, making it difficult to clearly capture the speech of attendees in the back rows. Secondly, most devices do not specifically optimize the sound wave transmission path. Microphones are usually placed on the frame or back of the device, and the display screen, casing, and other structures in front of them will reflect the sound waves, preventing the target sound energy from efficiently reaching the microphone diaphragm. Furthermore, in terms of acoustic environment adaptability, modern conference rooms are mostly made of glass curtain walls or hard decorations, resulting in long reverberation times. Traditional beamforming algorithms degrade in strong reverberation environments, easily capturing too much reflected sound, leading to blurred speech and reduced clarity. In addition, the acoustic hardware of existing devices is often independently designed or simply stacked, and the algorithm parameters are fixed or set based on ideal acoustic models, failing to consider the actual impact of the specific structure of the device itself. Therefore, this application proposes a far-field pickup and interference suppression system based on structural and acoustic co-optimization. Summary of the Invention

[0004] The purpose of this invention is to provide a far-field sound pickup and interference suppression system based on structural and acoustic co-optimization to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a far-field sound pickup and interference suppression system based on structural and acoustic synergistic optimization, comprising a sound guiding structure component, a noise isolation component, a microphone array, an acoustic processing module, and a system control unit, and further comprising a structural-acoustic synergistic module; The sound guiding structure component is integrated into the device body and is used to guide sound waves in the target direction to the microphone array; The noise isolation component is disposed between the microphone array and the device housing to attenuate structurally transmitted vibrations; The acoustic processing module is used to perform beamforming, noise suppression, and echo cancellation processing on the audio signals acquired by the microphone array. The structure-acoustic synergy module is configured to: acquire the geometric parameters of the sound-guiding structure component and / or the vibration isolation characteristic parameters of the noise isolation component, and dynamically adjust at least one algorithm parameter in the acoustic processing module according to the acquired parameters, so as to achieve synergistic optimization of structural characteristics and acoustic processing.

[0006] Preferably, the sound guiding structure component includes one or more combinations of sound guiding grooves, sound guiding hole arrays, low-resistance sound wave channels, and labyrinth acoustic structures; the sound guiding structure component is arranged on the front frame, top, or back area of ​​the device.

[0007] Preferably, the noise isolation component includes a suspended microphone array mounting base, a double-layer composite damping material, and an acoustically transparent foam layer; the suspended mounting base uses a low-stiffness elastic material to mechanically decouple the microphone array board from the main housing of the device.

[0008] Preferably, the microphone array is arranged as a linear array, a circular array, or a multi-subarray composite array; and the mounting plane of the microphone array has a preset upward or downward tilt angle relative to the horizontal plane so that the main beam direction of the array is aligned with the target pickup area.

[0009] Preferably, the structure-acoustic coordination module is used to adjust the suppression frequency band and intensity parameters of the anti-reverberation algorithm in the acoustic processing module based on the cavity volume parameters inside the sound guiding structure component.

[0010] Preferably, the structure-acoustic co-processing module is configured to: when a structural noise interference event is detected through audio signal analysis, dynamically calculate and improve the suppression threshold and strength of the noise suppression algorithm in the acoustic processing module for the corresponding frequency band signal based on the vibration isolation characteristic parameters of the noise isolation component.

[0011] Preferably, the system control unit further includes an environmental scene adaptation module for detecting the noise type and reverberation intensity of the current environment, and selecting different structure-acoustic cooperative strategies based on the detection results. The strategies include at least a narrow beam high suppression strategy for strong reverberation environments and a wide beam strong noise reduction strategy for high noise environments.

[0012] Preferably, the structure-acoustic co-modulation module dynamically adjusts the sensitivity threshold of the speech activity detection unit based on the sound wave transmission characteristics of the sound guiding structure component, so as to improve the endpoint detection performance in reverberant or low signal-to-noise ratio environments.

[0013] Preferably, the system control unit coordinates the beam pointing of the sound guiding structure components and the microphone array, as well as the parameters of the acoustic processing module, to form a closed-loop linkage, and automatically optimizes the overall sound pickup performance based on the sound source localization results or preset scenarios.

[0014] Compared with the prior art, the beneficial effects of the present invention are: The optimized sound guiding structure can more effectively collect and converge far-field sound energy. Combined with the directional gain of the array itself, it can increase the maximum effective sound pickup distance of the system by 30% to 80%, while improving the signal-to-noise ratio of the target speech signal. Noise isolation components cut off or significantly attenuate the transmission of structural vibrations to the microphone from the physical transmission path, reducing typical office noise interference such as typing and keyboard sounds by 40% to 70%. It breaks down the barriers between hardware and algorithms, making audio processing algorithms no longer fixed "black boxes," but capable of being finely optimized according to the specific structural characteristics of each device, thus improving the consistency and upper limit of product performance. It can automatically adapt to meeting room environments of different sizes and acoustic characteristics, reducing the need for manual adjustments by users. This allows the all-in-one conference machine to provide stable and clear sound pickup in various scenarios, expanding the application range and practical value of the equipment. Detailed Implementation

[0015] Exemplary embodiments will be described in detail below. The technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0016] This invention provides a technical solution: a far-field sound pickup and interference suppression system based on structural and acoustic synergistic optimization, including a sound guiding structure component, a noise isolation component, a microphone array, an acoustic processing module, and a system control unit, and also includes a structural-acoustic synergistic module; The sound guiding structure component is integrated into the device body and is used to guide sound waves in the target direction to the microphone array; The noise isolation component is disposed between the microphone array and the device housing to attenuate structurally transmitted vibrations; The acoustic processing module is used to perform beamforming, noise suppression, and echo cancellation processing on the audio signals acquired by the microphone array. The structure-acoustic synergy module is configured to: acquire the geometric parameters of the sound-guiding structure component and / or the vibration isolation characteristic parameters of the noise isolation component, and dynamically adjust at least one algorithm parameter in the acoustic processing module according to the acquired parameters, so as to achieve synergistic optimization of structural characteristics and acoustic processing.

[0017] Example 1: This example describes a specific implementation scheme for a sound guide groove integrated into the top bezel of a conference all-in-one machine and an upward-tilting microphone array in a collaborative design.

[0018] Inside the upper frame of the conference all-in-one machine, there is a narrow, elongated sound guide channel extending horizontally. The opening of this channel faces the front of the conference room, and its cross-section is a tapering horn shape. The inner surface is covered with acoustic absorbing material to reduce internal reflections. The end of the sound guide channel leads to a sealed cavity inside the device. A six-microphone array with a ring layout is mounted in this cavity via a floating mounting bracket, and the array plane has a preset upward tilt angle relative to the horizontal plane. The acoustic principle of this design is that human voices from the front of the conference room enter through the opening of the sound guide channel, are converged by the horn-shaped structure, and the sound energy is enhanced and more effectively transmitted to the microphone array inside the cavity. The upward tilt angle of the array allows the main beam direction of the array to be better aligned with the mouth height area of ​​the speaker who is standing or sitting, which is particularly beneficial for picking up the voice of speakers at a distance. At the software level, the structure-acoustic coordination module stores the length, cross-sectional area variation curve, and cavity volume parameters of the sound guide groove. During system initialization, the coordination module calculates the transfer function of the structure for sound waves of different frequencies based on these parameters, and initializes the weight vector of the beamforming algorithm accordingly, further optimizing its beam pattern based on the physical sound guide to form a dual focusing effect. At the same time, for specific frequency resonances that may be generated by the cavity, the coordination module will pre-set the parameters of the anti-reverberation algorithm to suppress the sound coloration caused by resonance. When the device is working, the system control unit can combine camera information or sound source localization results to fine-tune the beamforming pointing angle to complement the fixed physical sound guide direction, thereby achieving tracking and sound pickup of moving speakers.

[0019] Example 2: This example focuses on the specific implementation of a structural noise isolation component for suppressing touch and vibration interference.

[0020] In the all-in-one conference machine, the microphone array module is connected to the main body of the device through a multi-level vibration isolation system. The core of this system is a "floating mounting base", the main body of which is made of low-stiffness elastic material. The microphone array plate is "suspended" in the cavity by the column and has no rigid contact with the surrounding rigid shell. Between the back of the microphone array plate and the device shell, there is a double layer of composite damping material. The inner layer is a high loss factor viscoelastic material and the outer layer is a porous acoustic foam. In addition, high-density rubber feet are set in the area where the bottom of the device contacts the table. When the user touches the screen or taps the table, the vibration generated is first attenuated by the rubber feet. After the residual vibration is transmitted to the device shell, it is absorbed by a large amount by the double-layer damping material and converted into heat energy. The floating mounting base further prevents the vibration from being directly transmitted to the microphone array plate. The structure-acoustic co-modulation module stores the transmission loss curve characteristic parameters of the vibration isolation system. During actual sound pickup, the collaborative module monitors the signal energy in the low-frequency band in real time. When typical impact low-frequency noise is detected and the energy exceeds the dynamic threshold derived from the vibration isolation characteristics, it is judged as a structural noise interference event. At this time, the collaborative module will immediately notify the noise suppression algorithm to significantly increase the suppression strength of the corresponding frequency band in the next few hundred milliseconds, and may temporarily adjust the sidelobe suppression strategy of beamforming to avoid pointing the beam in the direction of the vibration source by mistake. Through this collaborative mechanism of "physical isolation as the main method and algorithm emergency response as the auxiliary method", the interference of touch and knocking noise on the voice signal can be minimized.

[0021] Example 3: This example details the specific process of how structural parameters drive acoustic algorithms to adapt to strong reverberation environments.

[0022] Assuming this system is installed in a conference room with a large glass wall, where the reverberation time is relatively long, after the system is powered on, the environmental scene adaptation module can determine that it is currently in a "strong reverberation scene" through initial acoustic detection or by utilizing built-in sensor information. The system control unit then triggers the corresponding collaborative optimization strategy. Under this strategy, the structure-acoustic collaborative module first reads the geometric parameters of the current device's sound-guiding structure. Since the sound-guiding structure itself has a certain directivity, it can naturally suppress some late reflections from the side walls and rear walls. Based on the directivity pattern of the sound-guiding structure, the collaborative module calculates an initial "spatial filter" template, and then instructs the beamforming algorithm to apply it to the physical direction. Building upon the directional nature of the signal, the beamwidth is further narrowed to form a directional pickup beam. Simultaneously, the coordination module estimates the reverberation frequency that may be aggravated by the standing wave effect based on the volume parameters of the acoustic cavity inside the device, and adjusts the key processing frequency band and attenuation factor of the subsequent anti-reverberation algorithm accordingly. In addition, under strong reverberation, the trailing of the speech signal can affect the accuracy of speech activity detection. The coordination module dynamically increases the tail decision threshold of VAD by combining the sound attenuation characteristics of the sound guiding structure, making the detection more accurate and avoiding misjudging the reverberant tail as valid speech, thereby ensuring the stability of the subsequent processing flow and maintaining excellent speech pickup clarity even in harsh acoustic environments.

[0023] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A far-field sound pickup and interference suppression system based on structural and acoustic co-optimization, comprising a sound-guiding structure component, a noise isolation component, a microphone array, an acoustic processing module, and a system control unit, characterized in that, It also includes a structure-acoustic co-modulation module; The sound guiding structure component is integrated into the device body and is used to guide sound waves in the target direction to the microphone array; The noise isolation component is disposed between the microphone array and the device housing to attenuate structurally transmitted vibrations; The acoustic processing module is used to perform beamforming, noise suppression, and echo cancellation processing on the audio signals acquired by the microphone array. The structure-acoustic synergy module is configured to: acquire the geometric parameters of the sound-guiding structure component and / or the vibration isolation characteristic parameters of the noise isolation component, and dynamically adjust at least one algorithm parameter in the acoustic processing module according to the acquired parameters, so as to achieve synergistic optimization of structural characteristics and acoustic processing.

2. The far-field sound pickup and interference suppression system based on structural and acoustic co-optimization according to claim 1, characterized in that: The sound guiding structure component includes one or more combinations of sound guiding grooves, sound guiding hole arrays, low-resistance sound wave channels, and labyrinth acoustic structures; the sound guiding structure component is arranged on the front frame, top, or back area of ​​the device.

3. The far-field sound pickup and interference suppression system based on structural and acoustic co-optimization according to claim 1, characterized in that: The noise isolation component includes a suspended microphone array mounting base, a double-layer composite damping material, and an acoustic transparent foam layer; the suspended mounting base uses a low-stiffness elastic material to mechanically decouple the microphone array board from the main housing of the device.

4. The far-field sound pickup and interference suppression system based on structural and acoustic co-optimization according to claim 1, characterized in that: The microphone array is arranged in the form of a linear array, a circular array, or a multi-subarray composite array; and the mounting plane of the microphone array has a preset upward or downward tilt angle relative to the horizontal plane so that the main beam direction of the array is aligned with the target sound pickup area.

5. The far-field sound pickup and interference suppression system based on structural and acoustic co-optimization according to claim 1, characterized in that: The structure-acoustic coordination module is used to adjust the suppression frequency band and intensity parameters of the anti-reverberation algorithm in the acoustic processing module based on the cavity volume parameters inside the acoustic guiding structure component.

6. The far-field sound pickup and interference suppression system based on structural and acoustic co-optimization according to claim 1, characterized in that: The structure-acoustic co-processing module is configured to: when a structural noise interference event is detected through audio signal analysis, dynamically calculate and improve the suppression threshold and strength of the noise suppression algorithm in the acoustic processing module for the corresponding frequency band signal based on the vibration isolation characteristic parameters of the noise isolation component.

7. The far-field sound pickup and interference suppression system based on structural and acoustic co-optimization according to claim 1, characterized in that: The system control unit also includes an environment scene adaptation module, which is used to detect the noise type and reverberation intensity of the current environment, and select different structure-acoustic cooperative strategies based on the detection results. The strategies include at least a narrow beam high suppression strategy for strong reverberation environment and a wide beam strong noise reduction strategy for high noise environment.

8. The far-field sound pickup and interference suppression system based on structural and acoustic co-optimization according to claim 1, characterized in that: The structure-acoustic co-modulation module dynamically adjusts the sensitivity threshold of the speech activity detection unit based on the sound wave transmission characteristics of the sound guiding structure components, thereby improving the endpoint detection performance in reverberant or low signal-to-noise ratio environments.

9. A far-field sound pickup and interference suppression system based on structural and acoustic co-optimization according to claim 1, characterized in that: The system control unit coordinates the beam pointing of the sound guiding structure components and the microphone array, as well as the parameters of the acoustic processing module, to form a closed-loop linkage, and automatically optimizes the overall sound pickup performance based on the sound source localization results or preset scenarios.