Wireless audio equipment low-power-consumption communication method and device, SoC chip and storage medium

Abstract

The invention relates to the technical field of wireless communication, and discloses a low-power-consumption communication method and device for wireless audio equipment, an SoC chip and a storage medium. The method comprises the following steps: sampling acoustic characteristics of heterogeneous audio data of a plurality of target audio devices, calculating a voice spectrum corresponding to device characteristics, and extracting multi-device acoustic characteristics, performing multi-frequency sub-band speech enhancement, heterogeneous segmentation and grouping of multi-target audio equipment, verification value calculation and differential coding on each extracted acoustic feature sequence to obtain a multi-source coding speech segment; performing transmission unit division and cooperative signal modulation, cooperative resource allocation and data stream transmission of each transmission unit on the multi-source coding voice segment to obtain a transmission voice data stream; and obtaining a transmission voice data stream transmitted with low power consumption and performing audio recovery on the obtained data to obtain a target audio data stream transmitted by the multi-target audio equipment. The communication power consumption of the multiple target audio devices is reduced, and the audio transmission quality is ensured.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to a low-power communication method and device for wireless audio equipment, a SoC chip and a storage medium. Background Art

[0002] In the field of wireless audio communication, low-power data transmission between audio devices is one of the key issues in current research. With the popularization of smart audio devices, the demand for audio data transmission between multiple devices is increasing, and the problem of device energy consumption is becoming increasingly prominent. At present, the industry generally adopts the method of compressing and encoding audio data before transmission to reduce energy consumption, or uses adaptive power control technology to adjust the transmission power to reduce transmission power consumption. However, these methods have many problems when dealing with the transmission of heterogeneous audio data from multiple devices, such as insufficient data feature extraction, severe signal distortion, high transmission energy consumption, etc., which are difficult to meet the needs of actual applications.

[0003] Nowadays, sub-band coding technology is used to divide the frequency of audio data, or a cooperative transmission scheme is used for resource scheduling in order to reduce the power consumption of audio data transmission. However, these methods still face challenges such as insufficient utilization of data features, poor coordination between devices, and inefficient resource allocation. Especially when dealing with data transmission of multiple heterogeneous audio devices, these methods often ignore some unique characteristics of audio data, such as differences in device characteristics, signal spectrum characteristics, data priority, etc., which have a significant impact on transmission energy consumption and audio quality. That is, the existing low-power communication methods for wireless audio devices ignore the multi-dimensional characteristics of audio data and the heterogeneous characteristics between devices, resulting in low energy utilization efficiency during transmission and difficulty in ensuring the transmission quality of audio data. Summary of the invention

[0004] The main purpose of the present invention is to solve the problem that the existing low-power communication method for wireless audio devices ignores the multi-dimensional characteristics of audio data and the heterogeneous characteristics between devices, resulting in low energy utilization efficiency during transmission and difficulty in ensuring the transmission quality of audio data.

[0005] A first aspect of the present invention provides a low-power communication method for a wireless audio device, the low-power communication method for a wireless audio device comprising: sampling acoustic features of heterogeneous audio data of multiple target audio devices to obtain multiple initial acoustic sampling points, and performing speech spectrum calculation of corresponding device characteristics on each of the initial acoustic sampling points to obtain frequency domain feature data corresponding to each of the target audio devices; extracting acoustic features of multiple devices from the frequency domain feature data to obtain multiple acoustic feature sequences, and performing speech enhancement of multiple frequency sub-bands on each of the acoustic feature sequences to obtain multiple enhanced speech streams; Perform heterogeneous segmentation grouping of each enhanced voice stream for multi-target audio devices to obtain voice stream grouping results, perform check value calculation and differential encoding on the voice stream grouping results to obtain multi-source coded voice segments; perform transmission unit division and collaborative signal modulation of each transmission unit on the multi-source coded voice segments to obtain modulated audio streams, perform collaborative resource allocation and data stream transmission on the modulated audio streams to obtain transmission voice data streams; obtain transmission voice data streams with low power consumption transmission, perform audio recovery on the transmission voice data streams to obtain target audio data streams transmitted by multi-target audio devices.

[0006] Optionally, in a first implementation method of the first aspect of the present invention, the sampling of acoustic features of heterogeneous audio data of multiple target audio devices to obtain multiple initial acoustic sampling points includes: based on a preset device sampling rate, downsampling the heterogeneous audio data of multiple target audio devices to obtain original audio data, and based on a preset fixed sampling interval, framing the original audio data to obtain initial audio data frames; performing sound pressure detection on the initial audio data frames to obtain audio energy distribution data, and performing threshold detection of voice activity on the audio energy distribution data to obtain voice segment data; performing continuity detection and timing marking on the voice segment data to obtain audio frame sequence data, and marking and sampling value quantization on the audio frame sequence data based on the device identification code corresponding to each of the target audio devices to obtain multiple initial acoustic sampling points.

[0007] Optionally, in a second implementation manner of the first aspect of the present invention, the voice spectrum calculation of the corresponding device characteristics of each of the initial acoustic sampling points to obtain frequency domain feature data corresponding to each of the target audio devices includes: performing mean and variance statistics on the initial acoustic sampling points to obtain an audio statistical sequence, and calculating a deviation threshold of the audio statistical sequence based on a preset confidence interval and audio statistical parameters; extracting sampling points whose sampling point amplitudes are greater than the deviation threshold from the initial acoustic sampling points to obtain valid audio sampling points, and calculating frequency domain transformation values ​​of each of the valid audio sampling points based on the corresponding frequency response coefficients of the target audio device to obtain an initial spectrum sequence; calculating the energy distribution of each of the initial spectrum sequences in a preset spectrum band, and filtering out frequency components greater than a preset device response interval to obtain a filtered spectrum sequence, and integrating the filtered spectrum sequence with the corresponding audio sampling moment and the device code corresponding to each of the target audio devices to obtain the frequency domain feature data corresponding to each of the target audio devices.

[0008] Optionally, in a third implementation method of the first aspect of the present invention, the extraction of multi-device acoustic features of the frequency domain feature data to obtain a multi-channel acoustic feature sequence includes: respectively calculating the energy distribution entropy values ​​of the frequency domain feature data in preset multiple time windows to obtain a frequency domain entropy value sequence, and calculating the frequency spectrum flatness of the frequency domain feature data to obtain a frequency domain flatness sequence, and extracting the modulation frequency component of the frequency domain feature data to obtain a frequency domain modulation feature sequence, and integrating parameters of the frequency domain entropy value sequence, frequency domain flatness sequence and frequency domain modulation feature sequence to construct an audio feature parameter table; performing a linear transformation on the feature parameter table to obtain a benchmark audio feature set, and performing a feature projection transformation on the frequency domain audio feature data to obtain an acoustic feature sequence; dividing the acoustic feature sequence into frequency bands to obtain a sub-band feature sequence, and performing energy normalization on the sub-band feature sequence to obtain a multi-channel acoustic feature sequence.

[0009] Optionally, in a fourth implementation method of the first aspect of the present invention, the speech enhancement of multiple frequency sub-bands on each of the acoustic feature sequences to obtain a multi-channel enhanced speech stream includes: estimating the sub-band noise power spectrum of the sub-band feature sequence in the acoustic feature sequence to obtain a sub-band noise spectrum, and calculating the signal power spectrum of the sub-band feature sequence to obtain a sub-band signal spectrum; based on the sub-band noise spectrum, performing spectral subtraction noise reduction on each sub-band feature in the sub-band signal spectrum to obtain a denoised sub-band signal; performing adaptive gain and phase compensation on the denoised sub-band signal to obtain an enhanced sub-band signal, and performing adaptive sub-band synthesis on each of the enhanced sub-band signals to obtain a multi-channel enhanced speech stream.

[0010] Optionally, in a fifth implementation of the first aspect of the present invention, the enhanced voice streams are heterogeneously segmented and grouped for multiple target audio devices to obtain voice stream grouping results, and the voice stream grouping results are checked and differentially encoded to obtain multi-source encoded voice segments, including: short-time energy calculation and zero-crossing rate calculation are performed on each enhanced voice stream to obtain an energy sequence and a zero-crossing rate sequence, and feature combination is performed on the energy sequence and the zero-crossing rate sequence to obtain device feature parameters; mutual correlation coefficients between the feature parameters of each device are calculated to generate a device correlation table, and based on the device correlation table, the audio feature distances between the target audio devices are calculated to generate device distances. matrix; constructing a hierarchical structure of the device distance matrix to obtain a device grouping structure, and indexing and assigning groups to the enhanced voice streams corresponding to each of the target audio devices in the device grouping structure to obtain a voice stream grouping result; dividing the voice stream grouping result into blocks to obtain a data block sequence, and averaging the data block sequences corresponding to different target audio devices in the same group to obtain common audio features; calculating the signal-to-noise ratio and weight normalizing the common audio features to obtain audio device weights, and based on the audio device weights, weighted combining the common audio features to obtain a fused feature vector; adaptively quantizing and coding checking the fused feature vector to obtain a multi-source coded speech segment.

[0011] Optionally, in a sixth implementation manner of the first aspect of the present invention, the multi-source coded voice segment is divided into transmission units and the coordinated signals of each transmission unit are modulated to obtain a modulated audio stream, and the modulated audio stream is subjected to coordinated resource allocation and data stream transmission to obtain a transmission voice data stream, including: calculating the amount of information and dividing the transmission levels of the multi-source coded voice segment to obtain an audio transmission priority, and based on the audio transmission priority, dynamically allocating bits and configuring a variety of coordinated transmission resource parameters for the multi-source coded voice segment to generate a modulated audio stream; calculating the modulation power consumption value of the preset modulation link when modulating the modulated audio stream, and performing segmented threshold division and resource quantity configuration calculation on the modulation power consumption value to obtain an initial resource allocation value; integrating and allocating a variety of modulation allocation parameters for the initial resource allocation value to generate a resource allocation parameter, and performing multi-device timing division and power level division on the resource allocation parameter to obtain a power configuration parameter; performing flow parameter calculation on the power configuration parameter to obtain a transmission data flow parameter, and based on the transmission data flow parameter and the resource allocation parameter, performing data stream transmission on the modulated audio stream to obtain a transmission voice data stream.

[0012] The second aspect of the present invention provides a low-power communication device for wireless audio devices, and the low-power communication device for wireless audio devices includes: a spectrum calculation module, which is used to sample acoustic features of heterogeneous audio data of multiple target audio devices to obtain multiple initial acoustic sampling points, and perform speech spectrum calculation of corresponding device characteristics on each of the initial acoustic sampling points to obtain frequency domain feature data corresponding to each of the target audio devices; a speech enhancement module, which is used to extract acoustic features of multiple devices from the frequency domain feature data to obtain multiple acoustic feature sequences, and perform speech enhancement of multiple frequency sub-bands on each of the acoustic feature sequences to obtain multiple enhanced speech streams; a segmentation grouping module A block is used to perform heterogeneous segmentation grouping of each enhanced voice stream for multi-target audio devices to obtain a voice stream grouping result, and perform check value calculation and differential encoding on the voice stream grouping result to obtain a multi-source coded voice segment; a signal modulation module is used to divide the multi-source coded voice segment into transmission units and perform collaborative signal modulation of each transmission unit to obtain a modulated audio stream, and perform collaborative resource allocation and data stream transmission on the modulated audio stream to obtain a transmission voice data stream; an audio recovery module is used to obtain a transmission voice data stream for low-power transmission, and perform audio recovery on the transmission voice data stream to obtain a target audio data stream transmitted by multi-target audio devices.

[0013] The third aspect of the present invention provides a SoC chip, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the SoC chip executes each step of the above-mentioned low-power communication method for wireless audio devices.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the various steps of the above-mentioned low-power communication method for wireless audio devices.

[0015] The above-mentioned low-power communication method, device, SoC chip and storage medium for wireless audio devices. In the embodiment of the present invention, acoustic features of heterogeneous audio data of multiple target audio devices are sampled to obtain multiple initial acoustic sampling points, and the voice spectrum of the corresponding device characteristics is calculated for each initial acoustic sampling point to obtain frequency domain feature data corresponding to each target audio device; the frequency domain feature data is extracted from the acoustic features of multiple devices to obtain multiple acoustic feature sequences, and each acoustic feature sequence is voice enhanced in multiple frequency sub-bands to obtain multiple enhanced voice streams; each enhanced voice stream is heterogeneously segmented and grouped for multiple target audio devices to obtain voice stream grouping results, and the voice stream grouping results are checked and differentially encoded to obtain multi-source coded voice segments; the multi-source coded voice segments are divided into transmission units and the coordinated signal modulation of each transmission unit is performed to obtain a modulated audio stream, and the modulated audio stream is subjected to coordinated resource allocation and data stream transmission to obtain a transmission voice data stream; the transmission voice data stream of low-power transmission is obtained, and the transmission voice data stream is audio restored to obtain the target audio data stream transmitted by the multiple target audio devices. Compared with the prior art, this application samples and performs spectrum analysis on heterogeneous data of multiple audio devices, obtains the frequency domain features of each device, then extracts the acoustic features of these frequency domain features, and performs voice enhancement through sub-band processing; then the enhanced voice stream is grouped and encoded; the encoded voice segment is divided into transmission units and signal modulated, and data is transmitted according to the resource allocation strategy; finally, the audio signal of the received data is restored. Through hierarchical data processing and transmission control, the problem of coordinated transmission of heterogeneous audio data of multiple devices is solved, especially in feature extraction and data fusion, the characteristic differences of different devices are fully considered, the data redundancy in the transmission process is effectively reduced, and the hierarchical modulation and resource allocation strategy are adopted, which not only ensures the audio quality, but also reduces the transmission power consumption, thereby realizing the overall efficient and low-power transmission of heterogeneous audio data.

[0016] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0017] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of a first embodiment of a low-power communication method for a wireless audio device according to an embodiment of the present invention; Figure 2The figure is a schematic diagram of an embodiment of a low-power communication device for wireless audio equipment in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] The terms "including" and "having" and any variations thereof mentioned in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device end including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or device ends.

[0021] To facilitate understanding of this embodiment, the specific process of the embodiment of the present invention is described below. Figure 1 , a first embodiment of a low-power communication method for a wireless audio device in an embodiment of the present invention includes: 101. Sampling acoustic features of heterogeneous audio data of multiple target audio devices to obtain multiple initial acoustic sampling points, and calculating the speech spectrum of the corresponding device characteristics for each initial acoustic sampling point to obtain frequency domain feature data corresponding to each target audio device; The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0022] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0023] In this embodiment, the target audio device here refers to various audio devices for audio data transmission, such as wireless headphones, smart speakers, microphone arrays, portable speakers, etc.; the heterogeneous audio data here refers to audio data from different types of audio devices with different characteristics (such as sampling rate, bit depth, number of channels and other parameters), for example: audio data collected by headphones: the sampling rate may be 48kHz, 16-bit quantization; data collected by the microphone array: may have multiple channels, the sampling rate is 16kHz, 24-bit quantization; audio data output by the speaker: may be 44.1kHz sampling rate, 32-bit quantization; the device characteristics here refer to the inherent characteristic differences in hardware performance and audio acquisition processing of different audio devices; based on a preset device sampling rate, the heterogeneous audio data of multiple target audio devices are downsampled to obtain original audio data, and based on a preset fixed sampling interval, the original audio data is framed to obtain initial audio data frames; sound pressure detection is performed on the initial audio data frames to obtain audio energy distribution data, and the audio energy distribution is analyzed. The method comprises the following steps: performing threshold detection of voice activity on the data of the distribution to obtain voice segment data; performing continuity detection and time sequence marking on the voice segment data to obtain audio frame sequence data, and marking and sampling value quantizing the audio frame sequence data based on the device identification code corresponding to each target audio device to obtain multiple initial acoustic sampling points; performing mean and variance statistics on the initial acoustic sampling points to obtain an audio statistical sequence, and calculating a deviation threshold of the audio statistical sequence based on a preset confidence interval and audio statistical parameters; extracting sampling points whose sampling point amplitudes are greater than the deviation threshold from the initial acoustic sampling points to obtain valid audio sampling points, and calculating the frequency domain transformation value of each valid audio sampling point based on the corresponding frequency response coefficient of the target audio device to obtain an initial spectrum sequence; calculating the energy distribution of each initial spectrum sequence in a preset spectrum frequency band, and filtering out frequency components greater than a preset device response interval to obtain a filtered spectrum sequence, and integrating the filtered spectrum sequence with the corresponding audio sampling moment and the device code corresponding to each target audio device to obtain frequency domain feature data corresponding to each target audio device.

[0024] In practical applications, firstly, preset sampling rate parameters are set according to the sampling capabilities of different audio devices, such as 48kHz for high-performance devices and 16kHz for low-performance devices, and then the heterogeneous audio data collected by each device are downsampled according to the lowest sampling rate standard, so as to unify the audio data of different devices to the same sampling rate standard and generate original audio data. Then, based on the preset fixed sampling interval (such as 20ms), these original audio data are framed and the continuous audio data are divided into data frames of equal length to form initial audio data frames. Then, by calculating the sound pressure value of each frame of data (that is, converting the audio signal amplitude into the sound pressure level in decibels), the audio energy distribution data reflecting the audio energy distribution characteristics of each frame is obtained, and the data is used as the initial audio data frame. Compare with the preset voice activity detection threshold (usually set in the range of 20-30dB), screen out valid voice segments that exceed the threshold, and generate voice segment data; then for the obtained voice segment data, by detecting the energy continuity and time relationship between adjacent frames, judge the integrity and timing relationship of the voice segment, and add a timestamp mark to each voice segment to generate audio frame sequence data with timing information, and finally add device identification information to the audio frame sequence data corresponding to each audio device according to the unique device identification code (such as device ID or MAC address) pre-assigned to each audio device, and quantize the sampling value according to the quantization accuracy of the device (such as 16-bit quantization), and finally obtain multiple initial acoustic sampling points with device identification and timing information. This realizes the orderly organization of multi-channel audio data and efficiently processes and transmits audio data of multiple devices.

[0025] Secondly, by calculating the mean (reflecting the DC bias level of the signal) and variance (reflecting the degree of fluctuation of the signal) of each sampling sequence in the initial acoustic sampling points, an audio statistical sequence describing the overall statistical characteristics of the audio signal is formed. Then, based on the pre-set confidence interval (usually a confidence level of 95% or 99%) and the obtained statistical parameters, the deviation threshold that can distinguish normal signals from abnormal values ​​is calculated using statistical theory. Then, each sampling value in the original initial acoustic sampling point is compared with the calculated deviation threshold, and the sampling points with amplitudes greater than the threshold are extracted (where these sampling points contain valid audio information), thereby obtaining valid audio sampling points. Then, according to the frequency response characteristics of different audio devices (including the gain response curve of the device at different frequencies), the threshold is determined for each audio device. The corresponding frequency response coefficients are applied to the frequency domain conversion process of the effective audio sampling points. The sampling points in the time domain are converted into the spectrum sequence in the frequency domain by fast Fourier transform and other methods to obtain the initial spectrum sequence. After obtaining the initial spectrum sequence, the energy distribution in each pre-set frequency band (such as low frequency band 20-200Hz, medium frequency band 200-2000Hz, high frequency band 2000-20000Hz) is calculated. At the same time, according to the effective frequency response range preset by each device (that is, the frequency range in which the device can work normally), the frequency components beyond this range are filtered out to obtain the filtered spectrum sequence. Finally, these filtered spectrum sequences are integrated with the time information recorded when the data is collected and the unique identification code of each device to form a complete frequency domain feature data packet. Not only the validity judgment and frequency domain conversion of audio data are realized, but also the frequency response characteristics of the device are considered to make the processing results more in line with the working characteristics of the actual device. At the same time, the redundancy of data transmission is reduced and the processing efficiency is improved by screening and integrating the frequency components.

[0026] 102. Extracting acoustic features of multiple devices from the frequency domain feature data to obtain a multi-channel acoustic feature sequence, and performing speech enhancement of multiple frequency sub-bands on each acoustic feature sequence to obtain a multi-channel enhanced speech stream; In this embodiment, the energy distribution entropy values ​​of the frequency domain feature data in the preset multiple time windows are calculated respectively to obtain the frequency domain entropy value sequence, the spectrum flatness calculation of the frequency domain feature data is calculated to obtain the frequency domain flatness sequence, and the modulation frequency component of the frequency domain feature data is extracted to obtain the frequency domain modulation feature sequence, and the frequency domain entropy value sequence, the frequency domain flatness sequence and the frequency domain modulation feature sequence are parameterized to construct an audio feature parameter table; the feature parameter table is linearly transformed to obtain a reference audio feature set, and the frequency domain audio feature data is feature projected to obtain an acoustic feature sequence; the acoustic feature The sequence is divided into frequency bands to obtain a sub-band feature sequence, and the sub-band feature sequence is energy normalized to obtain a multi-channel acoustic feature sequence; the sub-band noise power spectrum of the sub-band feature sequence in the acoustic feature sequence is estimated to obtain a sub-band noise spectrum, and the signal power spectrum of the sub-band feature sequence is calculated to obtain a sub-band signal spectrum; based on the sub-band noise spectrum, spectral subtraction is performed on each sub-band feature in the sub-band signal spectrum to obtain a denoised sub-band signal; adaptive gain and phase compensation is performed on the denoised sub-band signal to obtain an enhanced sub-band signal, and adaptive sub-band synthesis is performed on each enhanced sub-band signal to obtain a multi-channel enhanced speech stream.

[0027] In practical applications, we first calculate the energy distribution entropy of the frequency domain feature data in multiple preset time windows (such as analysis windows of different scales such as 10ms, 20ms, and 30ms) (where this entropy reflects the uncertainty and complexity of the signal in the frequency domain), thereby calculating the frequency domain entropy value sequence that describes the time-varying characteristics of the signal. At the same time, we perform spectrum flatness analysis on the frequency domain feature data, calculate the uniformity of energy distribution in each frequency band, and obtain the frequency domain flatness sequence that characterizes the spectral shape characteristics of the signal. We then extract the modulation features of the signal from the frequency domain feature data through methods such as bandpass filtering (where these modulation features reflect the periodic variation characteristics of the signal, forming a frequency domain modulation feature sequence). Then, we integrate the feature sequences of these three different dimensions (entropy value sequence, flatness sequence, and modulation feature sequence) according to the preset feature weights and combination rules. Construct an audio feature parameter table containing multi-dimensional feature information, then apply linear transformation processing (such as principal component analysis or linear discriminant analysis) to the feature parameter table, convert the original feature space to a new feature space, and obtain a more representative reference audio feature set. Use this feature set as a benchmark to perform feature projection transformation on the original frequency domain feature data, map high-dimensional features to low-dimensional feature space, and obtain a more compact acoustic feature sequence; then, according to the frequency characteristics of the audio signal, divide the entire frequency band into multiple sub-bands according to preset frequency boundaries (such as by critical band or octave division), process the features in each sub-band separately, and obtain a sub-band feature sequence. Finally, perform energy normalization processing on the feature sequences of each sub-band to eliminate the energy differences between different sub-bands, so that the features have better comparability, thereby obtaining a standardized multi-channel acoustic feature sequence. This method takes into account the multi-dimensional characteristics of the audio signal, and improves the expressiveness of the features through feature transformation and normalization processing, providing a feature basis for speech enhancement and data fusion.

[0028] Secondly, the noise power spectrum of the characteristic sequence in each sub-band is estimated. By analyzing the non-speech segment or low-energy segment of the signal, the background noise level in each sub-band is estimated to obtain the sub-band noise spectrum reflecting the noise characteristics. At the same time, the power spectral density analysis of the sub-band characteristic sequence is performed to calculate the energy distribution of the signal at each frequency point to obtain the sub-band signal spectrum representing the characteristics of the useful signal. Then, based on the sub-band noise spectrum, the signal spectrum and noise spectrum in each sub-band of the sub-band signal spectrum are spectrally subtracted, that is, the estimated noise spectrum is subtracted from the signal spectrum. This process needs to consider the balance between over-subtraction and under-subtraction, and is usually performed using a soft threshold method. Perform spectral subtraction to obtain the sub-band signal after preliminary noise reduction. Then, according to the energy distribution and signal-to-noise ratio of each sub-band signal, calculate the corresponding gain compensation coefficient, adjust the amplitude of the noise-reduced signal, and perform phase compensation based on the phase characteristics of the signal to eliminate the phase distortion that may be introduced by the spectral subtraction process, and obtain the enhanced sub-band signal with improved amplitude and phase. Finally, according to the importance and energy distribution of each sub-band signal, determine the appropriate sub-band synthesis weight, use the weighted overlap-add method to perform weighted synthesis of all enhanced sub-band signals, and reconstruct the complete speech signal. The weighted overlap-add formula is: ; in: is the weight coefficient of the kth subband, is the synthesis window function, is the subband signal, R is the frame shift, n is the time index or frame index, k is the subband or channel index, M is the length of the synthesis window function, and r is the sample index of the window function. For example, 8 subbands are set, adjacent frames overlap by 50%, and an improved triangular window is used for synthesis. Each subband is assigned a weight according to its signal-to-noise ratio, thereby obtaining a multi-channel enhanced speech stream with improved quality. The refined suppression of noise is achieved through subband processing, and the naturalness of the signal is ensured through gain and phase compensation. The final subband synthesis ensures the integrity of the enhanced speech, thereby reducing noise interference while maintaining the intelligibility and naturalness of the speech signal.

[0029] 103. Perform heterogeneous segment grouping of multiple target audio devices on each enhanced voice stream to obtain a voice stream grouping result, and perform check value calculation and differential coding on the voice stream grouping result to obtain a multi-source coded voice segment; In this embodiment, short-time energy calculation and zero-crossing rate calculation are performed on each enhanced voice stream to obtain an energy sequence and a zero-crossing rate sequence, and feature combination is performed on the energy sequence and the zero-crossing rate sequence to obtain device feature parameters; the mutual correlation coefficients between the feature parameters of each device are calculated to generate a device correlation table, and based on the device correlation table, the audio feature distances between the target audio devices are calculated to generate a device distance matrix; a hierarchical structure of the device distance matrix is ​​constructed to obtain a device grouping structure, and the enhanced voice streams corresponding to the target audio devices in the device grouping structure are indexed and grouped to obtain voice stream grouping results; the voice stream grouping results are divided into blocks to obtain a data block sequence, and the data block sequences corresponding to different target audio devices in the same group are averaged to obtain common audio features; the signal-to-noise ratio of the common audio features is calculated and weight normalized to obtain audio device weights, and based on the audio device weights, the common audio features are weighted combined to obtain a fused feature vector; the fused feature vector is adaptively quantized and coded to obtain a multi-source coded voice segment.

[0030] In practical applications, the energy sequence reflecting the change of signal strength is first obtained by calculating the signal energy value in each short-time frame, and the number of times the signal crosses the zero level in each analysis frame is calculated to obtain the zero-crossing rate sequence reflecting the frequency characteristics of the signal. Then, these two sequences are weighted fused according to the preset feature combination rules. The weighted fusion formula of the device characteristic parameters is: ; in, represents the energy of the nth frame, represents the zero-crossing rate, is the weighting coefficient, is the pitch feature, is the adjustment coefficient, N is the total number of frames, d is the identifier of the device, t is the index of the time period, and F(d) is the characteristic parameter value of device d in time period t, thereby calculating the device characteristic parameters that can characterize the audio characteristics of each device, and then calculating the mutual correlation coefficients between the characteristic parameters of different devices. These correlation coefficients reflect the similarity between the collected signals of different devices, thereby forming a device correlation table that describes the correlation between devices, and based on this correlation table, using measurement methods such as Euclidean distance or Mahalanobis distance to calculate the characteristic distance between each device, constructing a device distance matrix that characterizes the degree of difference between devices, and then applying hierarchical clustering and other methods to the distance matrix to construct a hierarchical structure, grouping devices with similar characteristic distances into a group, and forming a device correlation table. A grouping structure is prepared, and a unique index number is assigned to each group. The enhanced voice streams corresponding to each device are grouped and marked to obtain the voice stream grouping result. The grouped voice streams are then segmented according to a preset data block size to obtain a data block sequence that is easy to process. The mean of the data block sequences of different devices in the same group is calculated, and the common audio features reflecting the common characteristics of the group of devices are extracted. Then, the signal-to-noise ratio is calculated for these common features, and the quality of the collected signal of each device is evaluated. The signal-to-noise ratio value is converted into a weight after normalization processing to indicate the importance of different devices in feature fusion. Then, based on these weights, the common audio features are weighted combined to fuse the features of different devices in the same group. The weighted combination formula is: ; in, is the time domain characteristic of device i, is the frequency response function, is the variance, is the gain function, is the fusion weight, N is the total number of devices, t is the time index, and i is the index of the device, so as to calculate the fusion feature vector that can represent the overall characteristics of the group of devices. Finally, the fusion feature vector is fixed-point quantized and encoded with a checksum to obtain the final multi-source coded speech segment. This achieves the effective organization of heterogeneous data from multiple devices, ensures the rational use of the characteristics of each device, and prepares for subsequent low-power transmission through quantization coding, which not only maintains the characteristics of the signals of each device, but also achieves effective data fusion.

[0031] 104. Divide the multi-source coded voice segment into transmission units and perform coordinated signal modulation on each transmission unit to obtain a modulated audio stream, and perform coordinated resource allocation and data stream transmission on the modulated audio stream to obtain a transmission voice data stream; In this embodiment, the amount of information of multi-source coded voice segments is calculated and the transmission level is divided to obtain an audio transmission priority, and based on the audio transmission priority, dynamic bit allocation and multiple collaborative transmission resource parameters are configured for the multi-source coded voice segments to generate a modulated audio stream; the modulation power consumption value of the preset modulation link when modulating the modulated audio stream is calculated, and the modulation power consumption value is divided into segmented thresholds and resource amount configuration calculation is performed to obtain an initial resource allocation value; the initial resource allocation value is integrated and allocated with multiple modulation allocation parameters to generate a resource allocation parameter, and the resource allocation parameter is divided into power levels by multi-device timing to obtain a power configuration parameter; the power configuration parameter is calculated as a flow parameter to obtain a transmission data flow parameter, and based on the transmission data flow parameter and the resource allocation parameter, the modulated audio stream is transmitted as a data stream to obtain a transmission voice data stream.

[0032] In practical applications, the information entropy value of each coding segment is calculated to evaluate the amount of information. Then, these coding segments are divided into different transmission levels according to the distribution of the information, and the audio transmission priority that represents the importance of the data is obtained. Then, based on this priority, bits are allocated to the coded voice segments. More bit resources are allocated to high-priority data. At the same time, transmission resource parameters including modulation mode, coding mode and transmission power are configured. A modulated audio stream suitable for transmission is generated through the combination of these parameters. Then, the power consumption of each modulation link (including signal modulation, power amplification and channel coding, etc.) in the process of generating the modulated audio stream is calculated to obtain the modulation power consumption value reflecting the energy consumption. These power consumption values ​​are divided into different levels, and the initial resource allocation scheme is calculated in combination with the available resource conditions to obtain the initial resource allocation value. Then, this initial allocation value is integrated with multiple modulation parameters (such as modulation order, code rate, time slot length, etc.) to generate complete resource allocation parameters, and the transmission timing of multiple devices is planned according to these parameters. At the same time, the transmission power is divided into different levels to obtain detailed power configuration parameters, and then the data flow size of each transmission time slot is calculated based on the power configuration parameters to obtain the transmission data flow parameters used to control the transmission rate. Finally, these transmission parameters are applied to the actual transmission process of the modulated audio stream, and data is transmitted according to the allocated time slot, power and bit resources to obtain the final transmission voice data stream. Through multi-level parameter configuration and resource allocation, refined control of the transmission process is achieved, which not only ensures the transmission quality of important data, but also realizes the rational use of energy and low-power communication.

[0033] 105. Acquire a transmission voice data stream of low-power transmission, and perform audio recovery on the transmission voice data stream to obtain a target audio data stream transmitted by a multi-target audio device.

[0034] In this embodiment, the transmission voice data stream is prioritized to obtain hierarchical voice data, and the hierarchical voice data is divided into cache areas to obtain a data block location table; the number of accesses to the cache data in the data block location table is counted to obtain a data access record, and the importance of the data access record is calculated to obtain a replacement order table; the occupancy status of the cache data in the data block location table is detected to obtain a cache occupancy value, and the cache occupancy value is compared with a threshold to obtain cache screening data; the cache screening data is reorganized according to the replacement order table to obtain cache reorganized data, and the signal of the cache reorganized data is reconstructed to obtain a target audio data stream.

[0035] In actual applications, the target communication device receives the transmission voice data stream of low-power transmission, and then hierarchically processes the data stream according to the priority identification information carried in the data packet, divides the data of different importance into different priority categories, and obtains hierarchical voice data. Then, according to the available cache resources and the capacity requirements of data at each level, the corresponding cache areas are allocated to the data of different priorities, and the mapping relationship between the data blocks and the storage locations is established to form a data block location table. Then, the access frequency of the data blocks stored in each cache area is counted, and the number of times each data block is read and used is recorded, and a data access record reflecting the data usage is generated. Based on these access records, the importance value of each data block is calculated (wherein, this value comprehensively considers the priority and access frequency of the data, and is used to generate a guide cache data The cache area recorded in the data block location table is then checked for status, the data occupancy of each area is counted, and the cache occupancy value reflecting the cache usage status is obtained. These occupancy values ​​are compared with the preset cache capacity threshold, and the cache area that needs data replacement or merging is identified to obtain cache screening data. Then, the filtered cache data is reorganized according to the previously generated replacement sequence table, and the data with high importance is retained in the fast access area, while the data with low importance is moved to the spare area or the space occupied by it is released to obtain optimized cache reorganized data. Finally, the signal reconstruction processing is performed on the reorganized data, including re-splicing the segmented data, restoring the timing relationship of the signal, restoring the spectral characteristics of the audio signal, etc., and finally a complete target audio data stream is obtained. Therefore, through cache management and data reorganization, not only the fast access to important data is ensured, but also the efficient use of storage resources is achieved, and a stable and reliable communication data recovery function is realized.

[0036] In an embodiment of the present invention, the frequency domain features of each device are obtained by sampling and spectrum analysis of heterogeneous data of multiple audio devices, and then the acoustic features of these frequency domain features are extracted, and speech enhancement is performed through sub-band processing; then the enhanced voice stream is grouped and encoded; the encoded voice segment is divided into transmission units and signal modulated, and data is transmitted according to the resource allocation strategy; finally, the audio signal of the received data is restored. Through hierarchical data processing and transmission control, the problem of coordinated transmission of heterogeneous audio data of multiple devices is solved, especially in feature extraction and data fusion, the characteristic differences of different devices are fully considered, the data redundancy in the transmission process is effectively reduced, and the hierarchical modulation and resource allocation strategy are adopted, which not only ensures the audio quality but also reduces the transmission power consumption, thereby realizing the overall efficient and low-power transmission of heterogeneous audio data.

[0037] The low power consumption communication method of the wireless audio device in the embodiment of the present invention is described above. The low power consumption communication device of the wireless audio device in the embodiment of the present invention is described below. Figure 2 , an embodiment of a low-power communication device for wireless audio equipment in an embodiment of the present invention includes: The spectrum calculation module 201 is used to sample acoustic features of heterogeneous audio data of multiple target audio devices to obtain multiple initial acoustic sampling points, and perform speech spectrum calculation corresponding to device characteristics on each of the initial acoustic sampling points to obtain frequency domain feature data corresponding to each of the target audio devices; The speech enhancement module 202 is used to extract the acoustic features of multiple devices from the frequency domain feature data to obtain a multi-channel acoustic feature sequence, and perform speech enhancement of multiple frequency sub-bands on each of the acoustic feature sequences to obtain a multi-channel enhanced speech stream; The segmentation grouping module 203 is used to perform heterogeneous segmentation grouping of each enhanced voice stream for multiple target audio devices to obtain a voice stream grouping result, and perform check value calculation and differential coding on the voice stream grouping result to obtain a multi-source coded voice segment; The signal modulation module 204 is used to divide the multi-source coded voice segment into transmission units and perform coordinated signal modulation of each transmission unit to obtain a modulated audio stream, and perform coordinated resource allocation and data stream transmission on the modulated audio stream to obtain a transmission voice data stream; The audio recovery module 205 is used to obtain the transmission voice data stream of the low-power transmission, and perform audio recovery on the transmission voice data stream to obtain the target audio data stream transmitted by the multi-target audio device.

[0038] In an embodiment of the present invention, the frequency domain features of each device are obtained by sampling and spectrum analysis of heterogeneous data of multiple audio devices, and then the acoustic features of these frequency domain features are extracted, and speech enhancement is performed through sub-band processing; then the enhanced voice stream is grouped and encoded; the encoded voice segment is divided into transmission units and signal modulated, and data is transmitted according to the resource allocation strategy; finally, the audio signal of the received data is restored. Through hierarchical data processing and transmission control, the problem of coordinated transmission of heterogeneous audio data of multiple devices is solved, especially in feature extraction and data fusion, the characteristic differences of different devices are fully considered, the data redundancy in the transmission process is effectively reduced, and the hierarchical modulation and resource allocation strategy are adopted, which not only ensures the audio quality but also reduces the transmission power consumption, thereby realizing the overall efficient and low-power transmission of heterogeneous audio data.

[0039] above Figure 2 The low power consumption device of the wireless audio device in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The Soc chip in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0040] The Soc chip may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) (for example, one or more processors) and memories, and one or more storage media for storing applications or data (for example, one or more mass storage devices). Among them, the memory and the storage medium can be short-term storage or permanent storage. The program stored in the storage medium may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the Soc chip. Furthermore, the processor can be configured to communicate with the storage medium and execute a series of instruction operations in the storage medium on the Soc chip to implement the steps of the above-mentioned low-power consumption method for wireless audio devices.

[0041] The Soc chip may also have one or more wired or wireless network interfaces, one or more input and output interfaces, and / or one or more operating systems, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will appreciate that the Soc chip structure does not limit the Soc chip provided by the present invention, and may include more or fewer components than shown in the figure, or combine certain components, or arrange components differently.

[0042] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the low power consumption method for a wireless audio device.

[0043] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device, or unit can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0044] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0045] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A low-power communication method for wireless audio equipment, characterized in that: The low-power consumption communication method of the wireless audio device comprises: Sampling acoustic features of heterogeneous audio data of multiple target audio devices to obtain multiple initial acoustic sampling points, and calculating speech spectra of corresponding device characteristics for each of the initial acoustic sampling points to obtain frequency domain feature data corresponding to each of the target audio devices; Extracting acoustic features of multiple devices from the frequency domain feature data to obtain multi-channel acoustic feature sequences, and performing speech enhancement of multiple frequency sub-bands on each of the acoustic feature sequences to obtain multi-channel enhanced speech streams; Performing heterogeneous segment grouping of multiple target audio devices on each of the enhanced voice streams to obtain a voice stream grouping result, and performing check value calculation and differential coding on the voice stream grouping result to obtain a multi-source coded voice segment; Dividing the multi-source coded speech segment into transmission units and performing cooperative signal modulation on each transmission unit to obtain a modulated audio stream, and performing cooperative resource allocation and data stream transmission on the modulated audio stream to obtain a transmission voice data stream; A transmission voice data stream of low-power transmission is acquired, and audio recovery is performed on the transmission voice data stream to obtain a target audio data stream transmitted by a multi-target audio device.

2. The low-power communication method for wireless audio equipment according to claim 1, characterized in that: The sampling of acoustic features of heterogeneous audio data of multiple target audio devices to obtain multiple initial acoustic sampling points includes: Based on a preset device sampling rate, downsampling heterogeneous audio data of multiple target audio devices to obtain original audio data, and based on a preset fixed sampling interval, framing the original audio data to obtain initial audio data frames; Performing sound pressure detection on the initial audio data frame to obtain audio energy distribution data, and performing threshold detection of voice activity on the audio energy distribution data to obtain voice segment data; The voice segment data is subjected to continuity detection and timing marking to obtain audio frame sequence data, and based on the device identification code corresponding to each of the target audio devices, the audio frame sequence data is marked and the sampling value is quantized to obtain multiple initial acoustic sampling points.

3. The low-power communication method for wireless audio equipment according to claim 1, characterized in that: The performing of voice spectrum calculation of the corresponding device characteristics on each of the initial acoustic sampling points to obtain frequency domain feature data corresponding to each of the target audio devices includes: Performing mean and variance statistics on the initial acoustic sampling points to obtain an audio statistical sequence, and calculating a deviation threshold of the audio statistical sequence based on a preset confidence interval and audio statistical parameters; Extracting sampling points whose sampling point amplitudes are greater than the deviation threshold from the initial acoustic sampling points to obtain valid audio sampling points, and calculating frequency domain transformation values ​​of each of the valid audio sampling points based on corresponding frequency response coefficients of the target audio device to obtain an initial frequency spectrum sequence; The energy distribution of each of the initial spectrum sequences in a preset spectrum band is calculated, and the frequency components greater than the preset device response interval are filtered out to obtain a filtered spectrum sequence, and the filtered spectrum sequence is integrated with the corresponding audio sampling time and the device code corresponding to each of the target audio devices to obtain the frequency domain feature data corresponding to each of the target audio devices.

4. The low-power communication method for wireless audio equipment according to claim 1, characterized in that: The extracting of the acoustic features of multiple devices from the frequency domain feature data to obtain a multi-channel acoustic feature sequence includes: The energy distribution entropy values ​​of the frequency domain feature data in the preset multiple time windows are calculated respectively to obtain a frequency domain entropy value sequence, and the spectrum flatness of the frequency domain feature data is calculated to obtain a frequency domain flatness sequence, and the modulation frequency component of the frequency domain feature data is extracted to obtain a frequency domain modulation feature sequence, and the frequency domain entropy value sequence, the frequency domain flatness sequence and the frequency domain modulation feature sequence are parameterized to construct an audio feature parameter table; Performing a linear transformation on the feature parameter table to obtain a reference audio feature set, and performing a feature projection transformation on the frequency domain audio feature data to obtain an acoustic feature sequence; The acoustic feature sequence is divided into frequency bands to obtain sub-band feature sequences, and energy normalization is performed on the sub-band feature sequences to obtain multi-channel acoustic feature sequences.

5. The low-power communication method for wireless audio equipment according to claim 4, characterized in that: The step of performing speech enhancement of multiple frequency sub-bands on each of the acoustic feature sequences to obtain a multi-channel enhanced speech stream comprises: Performing sub-band noise power spectrum estimation on a sub-band feature sequence in the acoustic feature sequence to obtain a sub-band noise spectrum, and performing signal power spectrum calculation on the sub-band feature sequence to obtain a sub-band signal spectrum; Based on the sub-band noise spectrum, performing spectral subtraction on each sub-band feature in the sub-band signal spectrum to obtain a denoised sub-band signal; Adaptive gain and phase compensation are performed on the noise reduction sub-band signal to obtain an enhanced sub-band signal, and adaptive sub-band synthesis is performed on each of the enhanced sub-band signals to obtain a multi-channel enhanced voice stream.

6. The low-power communication method for wireless audio equipment according to claim 1, characterized in that: The step of performing heterogeneous segmentation grouping of each enhanced voice stream for multiple target audio devices to obtain a voice stream grouping result, and performing check value calculation and differential encoding on the voice stream grouping result to obtain a multi-source encoded voice segment includes: Performing short-time energy calculation and zero-crossing rate calculation on each of the enhanced speech streams to obtain an energy sequence and a zero-crossing rate sequence, and performing feature combination on the energy sequence and the zero-crossing rate sequence to obtain device feature parameters; Calculating the correlation coefficients between the characteristic parameters of each device to generate a device correlation table, and calculating the audio characteristic distances between the target audio devices based on the device correlation table to generate a device distance matrix; Constructing a hierarchical structure of the device distance matrix to obtain a device grouping structure, and performing index allocation grouping on the enhanced voice streams corresponding to each of the target audio devices in the device grouping structure to obtain a voice stream grouping result; The voice stream grouping result is divided into blocks to obtain a data block sequence, and the average of the data block sequences corresponding to different target audio devices in the same group is calculated to obtain a common audio feature; Calculating the signal-to-noise ratio and weight normalizing the common audio features to obtain audio device weights, and performing weighted combination on the common audio features based on the audio device weights to obtain a fused feature vector; Adaptive quantization and coding verification are performed on the fused feature vector to obtain a multi-source coded speech segment.

7. The low-power communication method for wireless audio equipment according to claim 1, characterized in that: The step of dividing the multi-source coded voice segments into transmission units and performing coordinated signal modulation on each transmission unit to obtain a modulated audio stream, and performing coordinated resource allocation and data stream transmission on the modulated audio stream to obtain a transmission voice data stream includes: Calculating the amount of information and dividing the transmission levels of the multi-source coded speech segments to obtain audio transmission priorities, and based on the audio transmission priorities, dynamically allocating bits and configuring multiple coordinated transmission resource parameters for the multi-source coded speech segments to generate a modulated audio stream; Calculating a modulation power consumption value of a preset modulation link when modulating the modulated audio stream, and performing segmented threshold division and resource amount configuration calculation on the modulation power consumption value to obtain an initial resource allocation value; Integrate and allocate multiple modulation allocation parameters for the initial resource allocation value to generate resource allocation parameters, and perform multi-device timing division and power level division on the resource allocation parameters to obtain power configuration parameters; A flow parameter calculation is performed on the power configuration parameter to obtain a transmission data flow parameter, and based on the transmission data flow parameter and the resource allocation parameter, data stream transmission is performed on the modulated audio stream to obtain a transmission voice data stream.

8. A low-power communication device for wireless audio equipment, characterized in that: The low-power consumption communication device of the wireless audio equipment comprises: A spectrum calculation module, used to sample acoustic features of heterogeneous audio data of multiple target audio devices to obtain multiple initial acoustic sampling points, and perform speech spectrum calculation of corresponding device characteristics on each of the initial acoustic sampling points to obtain frequency domain feature data corresponding to each of the target audio devices; A speech enhancement module, used to extract acoustic features of multiple devices from the frequency domain feature data to obtain a multi-channel acoustic feature sequence, and perform speech enhancement of multiple frequency sub-bands on each of the acoustic feature sequences to obtain a multi-channel enhanced speech stream; A segmentation grouping module is used to perform heterogeneous segmentation grouping of each enhanced voice stream for multiple target audio devices to obtain a voice stream grouping result, and perform check value calculation and differential coding on the voice stream grouping result to obtain a multi-source coded voice segment; A signal modulation module, used for dividing the multi-source coded voice segment into transmission units and performing coordinated signal modulation of each transmission unit to obtain a modulated audio stream, and performing coordinated resource allocation and data stream transmission on the modulated audio stream to obtain a transmission voice data stream; The audio recovery module is used to obtain the transmission voice data stream of low-power transmission and perform audio recovery on the transmission voice data stream to obtain the target audio data stream transmitted by the multi-target audio device.

9. A SoC chip, characterized in that: The SoC chip comprises: a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instructions in the memory to enable the SoC chip to execute each step of the low-power communication method for a wireless audio device as described in any one of claims 1 to 7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the steps of the low-power communication method for a wireless audio device as described in any one of claims 1 to 7 are implemented.

Images