A method and device for audio noise reduction of wireless headphones based on Polar code
By using Polar code and minimum energy encoding technology in wireless headphones, the audio data is encoded and equalized, and the diffusion model is used to reduce noise, the problem of high energy consumption in wireless headphone noise reduction function is solved, and a longer Battery life and higher quality audio signals.
Patent Information
- Application Number
- CN202411271580.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-09-11
AI Technical Summary
Existing wireless headphones have problems with high energy consumption and high computing resources in the noise reduction function, which affects device performance and battery life.
Using the audio noise reduction method based on Polar code, the audio data is encoded by obtaining the working mode of wireless headphones, using Polar code or minimum energy encoding, and performing minimum mean square error equalization processing and diffusion model noise reduction, optimized encoding and decoding processes to reduce energy consumption.
It effectively reduces the energy consumption of audio data transmission, extends the battery life of the device, and improves the quality and noise immunity of audio signals.
Smart Images

Figure CN118803502B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of wireless communications, and in particular relates to an audio noise reduction method and device for wireless headphones based on Polar codes. Background Art
[0002] In the field of modern communications and personal audio devices, wireless headphones have become one of the most widely used devices, not only because of the wireless convenience they provide, but also because of their increasingly enhanced audio processing capabilities, especially noise reduction in voice calls and music playback. However, with the increase in functionality, how to maintain or improve device performance without sacrificing battery life has become a major challenge in designing wireless headphones. Noise reduction technology is particularly important in wireless headphones because it directly affects the quality of calls and the clarity of music playback. Adopting efficient noise reduction algorithms can significantly improve the user experience, especially in noisy environments. However, these algorithms usually require high computing resources and energy consumption. Summary of the invention
[0003] In order to reduce the high energy consumption and complexity of the wireless headset noise reduction algorithm, a first aspect of the present invention provides an audio noise reduction method for wireless headsets based on Polar codes, including: obtaining the working mode of the wireless headset, and encoding the audio data through Polar codes or minimum energy coding according to different working modes; performing minimum mean square error equalization processing on the encoded audio data; for audio data with a minimum mean square error higher than a threshold, denoising the audio data through a diffusion model.
[0004] In some embodiments of the present invention, the audio data is encoded by Polar code or minimum energy coding according to different working modes: if the wireless headset is in voice mode, the audio data is encoded by Polar code optimized by LSTM model; if the wireless headset is in music mode, the audio data is encoded by Polar code; if the wireless headset is in energy saving mode, the audio data is encoded by minimum energy coding.
[0005] Furthermore, encoding the audio data using the Polar code optimized by the LSTM model includes: grouping the audio data into bit streams to generate an initial bit stream; optimizing the generation matrix of the Polar code through the LSTM model; and encoding the initial bit stream based on the optimized generation matrix.
[0006] Furthermore, encoding the audio data by minimum energy coding includes: mapping each information bit of the audio data to multiple codewords; and selecting a codeword mapping with minimum energy to encode the audio data.
[0007] In some embodiments of the present invention, the minimum mean square error equalization processing of the encoded audio data includes: obtaining the channel gain and noise power of the audio signal; calculating the minimum mean square error equalization coefficient of the optimized audio data based on the channel gain and the noise power; and performing equalization processing on the optimized audio data according to the minimum mean square error equalization coefficient.
[0008] In some embodiments of the present invention, the denoising of the audio data through a diffusion model includes: obtaining noise data of wireless headphones and training a diffusion model based on the noise data; generating initial noise of the audio data based on the trained diffusion model; and denoising the audio data based on reverse diffusion of the initial noise.
[0009] The second aspect of the present invention provides an audio noise reduction device for wireless headphones based on Polar codes, including: an encoding module, used to obtain the working mode of the wireless headphones, and encode audio data through Polar codes or minimum energy coding according to different working modes; an equalization module, used to perform minimum mean square error equalization processing on the encoded audio data; and a noise reduction module, used to perform noise reduction on the audio data with a minimum mean square error higher than a threshold through a diffusion model.
[0010] Furthermore, the encoding module includes: a first encoding module, which is used to encode the audio data through the Polar code optimized by the LSTM model if the wireless headset is in voice mode; a second encoding module, which is used to encode the audio data through the Polar code if the wireless headset is in music mode; and a third encoding module, which is used to encode the audio data through minimum energy coding if the wireless headset is in energy-saving mode.
[0011] In a third aspect of the present invention, an electronic device is provided, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the audio noise reduction method for wireless headphones based on Polar codes provided in the first aspect of the present invention.
[0012] A fourth aspect of the present invention provides a computer-readable medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the audio noise reduction method for wireless headphones based on Polar codes provided in the first aspect of the present invention is implemented.
[0013] The beneficial effects of the present invention are:
[0014] 1. Through minimum energy coding, the energy consumption in audio data transmission can be effectively reduced, which is suitable for low-power devices such as battery-powered wireless headphones. The implementation steps include data preprocessing, minimum energy coding, transmission and decoding, ensuring that the battery life of the device is maximized while maintaining audio quality;
[0015] 2. Perform minimum mean square error equalization on the encoded audio data to reduce the impact of channel noise and restore high-quality audio signals. The MMSE equalizer provides an efficient channel equalization method by minimizing the mean square error between the received signal and the transmitted signal, which is suitable for low-power devices such as wireless headphones. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 A schematic diagram of a basic flow chart of an audio noise reduction method for a wireless headset based on Polar codes in some embodiments of the present invention;
[0017] Figure 2 It is a schematic diagram of a specific process of an audio noise reduction method for a wireless headset based on Polar code in some embodiments of the present invention;
[0018] Figure 3 Schematic diagram of the structure of an audio noise reduction device for wireless headphones based on Polar codes in some embodiments of the present invention;
[0019] Figure 4 It is a schematic diagram of the structure of an electronic device in some embodiments of the present invention. DETAILED DESCRIPTION
[0020] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.
[0021] refer to Figure 1 and Figure 2 In a first aspect of the present invention, a method for audio noise reduction of a wireless headset based on Polar code is provided, comprising:
[0022] S100. Obtaining the working mode of the wireless headset, and encoding the audio data by Polar code or minimum energy coding according to different working modes;
[0023] S200. Perform minimum mean square error equalization on the encoded audio data;
[0024] S300. For audio data whose minimum mean square error is higher than a threshold, denoise the audio data using a diffusion model.
[0025] It can be understood that the basic working modes of wireless headphones include voice mode, music mode, and energy-saving mode (low power mode); in order to ensure effective energy saving of audio quality during the noise reduction process, multi-level noise reduction modes are selected or adapted according to different modes and user needs.
[0026] Therefore, in step S100 of some embodiments of the present invention, the audio data is encoded by Polar code or minimum energy coding according to different working modes:
[0027] S101. If the wireless headset is in voice mode, the audio data is encoded using the Polar code optimized by the LSTM model;
[0028] S102. If the wireless headset is in music mode, the audio data is encoded using Polar code;
[0029] S103. If the wireless headset is in the energy-saving mode, the audio data is encoded by minimum energy encoding.
[0030] Furthermore, in step S101, encoding the audio data by using the Polar code optimized by the LSTM model includes:
[0031] S1011. Group the audio data into bit streams to generate an initial bit stream;
[0032] Specifically, the audio data needs to be preprocessed to convert it into a bit stream suitable for encoding. That is, the audio data is converted into PCM (pulse code modulation) format; then the PCM data is quantized to generate a bit stream; and then the bit stream is grouped according to the input format of the Polar code.
[0033] S1012. Optimize the generation matrix of the Polar code through the LSTM model;
[0034] Specifically, Polar code is a coding method based on channel polarization phenomenon, which generates coded bits by arranging information bits and frozen bits (all 0 or all 1). That is, setting the parameters of Polar code, such as code length N and the number of information bits K Then the frozen bit positions are generated, and finally the information bits and frozen bits are arranged to generate the initial bit stream u. Then train the LSTM model to optimize the generator matrix of the Polar code. LSTM model training steps: Prepare training data, including input bit streams under different channel conditions and the corresponding Polar code encoded bit streams. Build an LSTM model, including input layer, hidden layer, and output layer, and train the model using loss functions (such as mean square error) and optimization algorithms (such as Adam). Save the trained model parameters. Optionally, other deep learning methods can be used to optimize Polar codes, such as GRU, CNN, RNN, etc.
[0035] S1013. Encode the initial bit stream based on the optimized generation matrix.
[0036] Specifically, the trained LSTM model is used to optimize the generator matrix of the Polar code. Steps for optimizing the generator matrix: u Input LSTM model; generate optimized generation matrix through LSTM model G LSTM . Then use the optimized generator matrix to encode and generate the coded bits x .
[0037] The initial bit stream u =[ u 1 ,u 2 ,…,u N ]; The optimized generator matrix G LSTM= LSTM(u) , coded bits x: x=u·G LSTM .
[0038] It can be understood that minimum energy coding maps each information bit to a codeword to reduce energy consumption. k The bit information is mapped to an n-bit codeword containing at most one high bit. Therefore, in step S103, encoding the audio data by minimum energy coding includes:
[0039] S1031. Map each information bit of the audio data into multiple codewords;
[0040] Specifically, the audio data is converted into a PCM (Pulse Code Modulation) format, the PCM data is quantized, and a bit stream is generated, and the bit stream is grouped according to the input format of the minimum energy coding.
[0041] S1032. Select a codeword mapping with minimum energy to encode the audio data.
[0042] in, represents the probability of minimum energy coding, n Indicates the codeword length, k Indicates the information bit length, P eM It indicates the probability that a high position is misjudged as a low position. P eS Indicates the probability of a low position being misjudged as a high position.
[0043] Accordingly, during transmission and decoding, the encoded bit stream needs to be modulated and converted into a wireless signal. During decoding, the ME codeword is decoded into information bits through demodulation and minimum energy decoding.
[0044] It can be understood that through minimum energy coding, the energy consumption in audio data transmission can be effectively reduced, which is suitable for low-power devices such as battery-powered wireless headphones. The implementation steps include data preprocessing, minimum energy coding, transmission and decoding, ensuring that the battery life of the device is maximized while maintaining audio quality.
[0045] Without loss of generality, in multi-level coding, information bits are divided into different levels for coding processing to achieve efficient coding and transmission. Therefore, based on the above, different requirements may be placed on the reliability and efficiency of data transmission according to different channel conditions (such as signal-to-noise ratio, fading characteristics). Layering according to channel conditions can ensure efficient coding and transmission in various environments. First level: For high signal-to-noise ratio conditions, simple coding methods (such as uncoded or low-complexity ME coding) are used to reduce computation and energy consumption. Second level: For medium signal-to-noise ratio conditions, medium-complexity coding methods (such as Polar codes) are used to balance coding efficiency and anti-interference ability. Third level: For low signal-to-noise ratio conditions, complex coding methods (such as deep learning optimized Polar codes) are used to maximize anti-interference ability and data reliability.
[0046] In addition, in wireless headset communications, different types of data (such as audio, control signals) may have different priorities. Layering according to data priority can ensure that high-priority data is transmitted and protected first. First layer: high-priority data (such as control signals) uses the most reliable encoding method (such as deep learning optimized Polar code) to ensure accurate data transmission. Second layer: medium-priority data (such as voice data) uses medium-complexity encoding methods (such as Polar codes) to balance transmission efficiency and data reliability. Third layer: low-priority data (such as background sound) uses low-complexity encoding methods (such as ME encoding) to reduce energy consumption.
[0047] Assuming the information bit u Divided into L Layer, each layerl The coded bits are u l , the total coded bits are x :
[0048] ,
[0049] in, C l Indicates l The encoding function of the layer.
[0050] In step S200 of some embodiments of the present invention, performing minimum mean square error equalization processing on the encoded audio data includes:
[0051] S201. Obtaining the channel gain and noise power of the audio signal;
[0052] Specifically, before performing MMSE equalization, it is necessary to first understand the characteristics of the channel and the model of the received signal. Generally speaking, the received signal can be expressed as:
[0053] ,
[0054] in, y Indicates receiving signal, h represents the channel gain, x Indicates the transmission signal. n The goal of the MMSE equalizer is to equalize the received signal y Converted to the estimated transmitted signal , to minimize the mean square error. The output of the MMSE equalizer can be expressed as:
[0055] ,
[0056] in, is the equalized signal, h H is the channel gain h The conjugate transpose of | h | 2 is the square of the channel gain, σ 2 is the noise power.
[0057] In the diffusion model, the signal condition after equalization is:
[0058] ,
[0059] Among them, the noise term after equalization Compared with the noise term before equalization, it has a smaller impact, making the equalized signal closer to the original transmitted signal. y eq The conditional distribution of can be expressed as: For Rayleigh fading channel, it is known that x and h The conditional distribution of the balanced signal is:
[0060] ,
[0061] in, , σ 2 is the noise power.
[0062] S202. Based on the channel gain and the noise power, calculating the minimum mean square error equalization coefficient of the optimized audio data;
[0063] Specifically, the goal of the MMSE equalizer is to recover the transmitted signal by minimizing the mean square error x The calculation formula of MMSE equalization coefficient is:
[0064] ,
[0065] in, h H is the channel gain h The conjugate transpose of | h | 2 is the square of the channel gain, σ 2 is the noise power.
[0066] S203. Perform equalization processing on the optimized audio data according to the minimum mean square error equalization coefficient. y Apply equalization coefficients to restore optimized audio data .
[0067] It can be understood that the diffusion model is a powerful generative model that can gradually remove noise and restore the original signal.
[0068] Therefore, in some embodiments of the present invention, the denoising of the audio data by using a diffusion model includes:
[0069] S301. Acquire noise data of the wireless headset, and train a diffusion model based on the noise data;
[0070] Specifically, the audio data is preprocessed and converted into a format suitable for diffusion model processing. The audio data is converted into PCM (Pulse Code Modulation) format; the PCM data is normalized so that its value is between [-1, 1].
[0071] It can be understood that the diffusion model gradually adds noise to the data in the forward diffusion process until the data becomes completely noisy. In the reverse diffusion process, the model learns to gradually remove the noise and restore the original data. Forward diffusion process:
[0072] ,
[0073] in: x t Indicates t The diffusion data of the step, x 0 means initial data (original audio), α t represents the attenuation factor in the forward process, represents the noise of standard normal distribution;
[0074] The reverse diffusion process model is constructed by learning x t recover x 0, that is:
[0075] ,
[0076] in, represents the noise in the model predictions.
[0077] The training process of the diffusion model includes the following steps: data preparation: prepare a training data set, including noisy audio data and corresponding noise-free audio data; model construction: build a diffusion model, including forward and reverse processes; training model: use the noisy audio data to train the model and learn the noise distribution.
[0078] S302. Based on the trained diffusion model, generate initial noise of the audio data;
[0079] S303: Based on the reverse diffusion of the initial noise, perform noise reduction on the audio data.
[0080] Specifically, generate the initial noisy data x T . The model is used to gradually remove noise and restore the original audio data.
[0081] It can be understood that a power consumption management module can be added to the above system to monitor and optimize the energy consumption during encoding and decoding in real time to ensure the battery life of the wireless headset.
[0082] Assume that the total power consumption of the system is P total , including encoding power consumption, decoding power consumption and other power consumption, then: P total = P enc+ P dec+ P other ;
[0083] in P total Indicates the total system power consumption, P enc represents the encoding power consumption, P dec represents the decoding power consumption, P other Represents other power consumption. Use low-power encoding technology and efficient decoding models to reduce the power consumption of encoding and decoding. Monitor system power consumption in real time and dynamically adjust encoding and decoding strategies to achieve optimal power management. Therefore, combined with multi-level encoding, multi-stage decoding technology and power management modules, we are able to achieve an efficient communication system with low power consumption, low latency and low bit error rate in wireless headset applications. By optimizing the encoding and decoding process and managing power consumption in real time, we ensure the stability and efficiency of the system under different channel conditions, meeting the needs of modern wireless headsets for high performance, long battery life and high-quality audio.
[0084] refer to Figure 3 In a second aspect of the present invention, there is provided an audio noise reduction device 1 for wireless headphones based on Polar codes, comprising:
[0085] The encoding module 11 is used to obtain the working mode of the wireless headset and encode the audio data through Polar code or minimum energy coding according to different working modes;
[0086] An equalization module 12, used for performing minimum mean square error equalization processing on the encoded audio data;
[0087] The noise reduction module 13 is used to reduce the noise of the audio data with a minimum mean square error higher than a threshold value by using a diffusion model.
[0088] Furthermore, the encoding module 11 includes:
[0089] A first encoding module, used for encoding audio data by using Polar code optimized by LSTM model if the wireless headset is in voice mode;
[0090] A second encoding module, configured to encode the audio data using a Polar code if the wireless headset is in a music mode;
[0091] The third encoding module is used to encode the audio data through minimum energy encoding if the wireless headset is in the energy saving mode.
[0092] refer to Figure 4 According to a third aspect of the present invention, an electronic device is provided, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the audio noise reduction method for wireless headphones based on Polar codes according to the first aspect of the present invention.
[0093] The electronic device 500 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage device 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.
[0094] Typically, the following devices may be connected to the I / O interface 505: an input device 506 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 507 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 508 including, for example, a hard disk, etc.; and a communication device 509. The communication device 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 4 The electronic device 500 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 4 Each block shown in the figure may represent one device, or may represent multiple devices as required.
[0095] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through a communication device 509, or installed from a storage device 508, or installed from a ROM 502. When the computer program is executed by the processing device 501, the above functions defined in the method of the embodiment of the present disclosure are executed. It should be noted that the computer-readable medium described in the embodiment of the present disclosure can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection with one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In an embodiment of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, an apparatus, or a device. In an embodiment of the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, an apparatus, or a device. The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wire, optical cable, RF (radio frequency), etc., or any suitable combination of the foregoing.
[0096] The computer-readable medium may be included in the electronic device, or may exist independently without being installed in the electronic device. The computer-readable medium carries one or more computer programs. When the one or more programs are executed by the electronic device, the electronic device:
[0097] Computer program code for performing the operations of embodiments of the present disclosure may be written in one or more programming languages or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, Python, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on a user's computer, partially on a user's computer, as a separate software package, partially on a user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0098] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0099] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. An audio noise reduction method for wireless headphones based on Polar codes, characterized in that: include: Obtain the working mode of the wireless headset, and encode the audio data using Polar code or minimum energy coding according to different working modes: If the wireless headset is in voice mode, encode the audio data using Polar code optimized by the LSTM model; If the wireless headset is in music mode, the audio data is encoded using Polar code; If the wireless headset is in energy-saving mode, the audio data is encoded using minimum energy coding; Perform minimum mean square error equalization processing on the encoded audio data; For audio data whose minimum mean square error is higher than a threshold, noise reduction is performed on the audio data using a diffusion model.
2. The method for audio noise reduction of wireless headphones based on Polar codes according to claim 1, characterized in that: The encoding of audio data by using the Polar code optimized by the LSTM model includes: Grouping the audio data into bit streams to generate an initial bit stream; The generation matrix of Polar code is optimized through LSTM model; The initial bit stream is encoded based on the optimized generator matrix.
3. The method for audio noise reduction of wireless headphones based on Polar codes according to claim 1, characterized in that: The encoding of the audio data by minimum energy coding comprises: Mapping each information bit of the audio data into a plurality of code words; A codeword mapping with minimum energy is selected to encode the audio data.
4. The method for audio noise reduction of wireless headphones based on Polar codes according to claim 1, characterized in that: The minimum mean square error equalization processing of the encoded audio data comprises: Obtain channel gain and noise power of the audio signal; Based on the channel gain and the noise power, calculating a minimum mean square error equalization coefficient of the optimized audio data; The optimized audio data is equalized according to the minimum mean square error equalization coefficient.
5. The method for audio noise reduction of wireless headphones based on Polar codes according to claim 1, characterized in that: The denoising of the audio data by using a diffusion model comprises: Acquire noise data of the wireless headset, and train a diffusion model based on the noise data; Based on the trained diffusion model, generating initial noise of the audio data; The audio data is subjected to noise reduction based on the reverse diffusion of the initial noise.
6. An audio noise reduction device for wireless headphones based on Polar codes, characterized in that: include: The encoding module is used to obtain the working mode of the wireless headset and encode the audio data through Polar code or minimum energy coding according to different working modes: if the wireless headset is in voice mode, the audio data is encoded through Polar code optimized by LSTM model; If the wireless headset is in music mode, the audio data is encoded using Polar code; If the wireless headset is in energy-saving mode, the audio data is encoded using minimum energy coding; An equalization module, used for performing minimum mean square error equalization processing on the encoded audio data; The denoising module is used to denoise the audio data whose minimum mean square error is higher than a threshold value by using a diffusion model.
7. An electronic device comprising: one or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, enables the one or more processors to implement the audio noise reduction method for wireless headphones based on Polar codes as described in any one of claims 1 to 5.
8. A computer readable medium having a computer program stored thereon, wherein: When the computer program is executed by a processor, the audio noise reduction method for a wireless headset based on Polar codes is implemented as described in any one of claims 1 to 5.
Citation Information
Patent Citations
High data rate transmission using minimum energy coding wth ultra wide band modulation
CN113395227A
Communication method and device
CN116195194A
Channel coding and modulation
CN116235534A
Bluetooth communication method and system, and electronic device
WO2022188813A1
