1884 results about "Sound quality" patented technology

The invention relates to the technical field of audio signal processing, and discloses a high-sampling-rate audio analysis optimization method and system, a storage medium and equipment, and the method comprises the steps: upgrading an FFTW (Fast Fourier Transform) library, and building an optimized memory management mechanism; carrying out adaptive framing processing on the input audio data, and optimizing the data access efficiency by adopting a multi-stage cache system; the calculation precision of the FFT fast Fourier transform is dynamically adjusted, and the calculation of the FFT fast Fourier transform is optimized by pre-calculating and caching an FFTW plan; performing clock synchronization optimization and jitter elimination on the COAX digital audio signal; establishing a multi-dimensional result cache matrix and an intelligent multiplexing decision tree, and reducing repeated operation through intelligent cache prediction and similarity calculation; system performance indexes are monitored in real time, adaptive parameter adjustment is realized based on an optimization objective function, audio quality and system stability are monitored in combination with subjective and objective evaluation mechanisms, repeated operation is remarkably reduced, and calculation resource allocation is optimized on the premise of ensuring sound quality.

The invention relates to the technical field of audio pushing optimization, and discloses an audio stream pushing crash prevention method and system, a storage medium and equipment, and the method comprises the steps: dynamically generating a format perception pre-detection thread based on a current audio stream format, and adjusting the priority and sampling frequency of the thread according to the audio bit depth and the sampling rate; multi-level intelligent buffer area management with sound quality priority is implemented, and dynamic adjustment of the size of a buffer area and adaptive switching of the buffer area are realized according to different audio formats and input source types; establishing a cross-brand equipment coordination framework, and setting a watchdog timer with preferential tone quality to perform timed monitoring so as to process response timeout; triggering an elegant degradation strategy of sound quality priority according to the resource availability score, and reducing the audio processing quality step by step to maintain the playing continuity; the key characteristics of each transmission protocol are identified and utilized, the transmission protocols are adaptively switched according to the network condition and the processing capability of brand equipment, and real high-quality and zero-interruption audio experience is brought to users.

The invention discloses a multi-device high-resolution audio stream intelligent pushing method and device, equipment and a storage medium, which are applied to a multi-device audio system, and the method comprises the following steps: acquiring a current audio stream format and adjusting system working parameters; dynamically adjusting a multi-stage buffer area based on a tone quality priority strategy; operation parameters of multiple devices are synchronously monitored in real time, and data are sent in advance for the device with large delay based on the delay characteristic and the synchronization capability of each device in the system; and dynamically switching system service levels based on the quality of the current audio stream. A protection strategy can be flexibly adjusted according to different audio formats through a format perception pre-detection thread mechanism; the sound quality is kept to the greatest extent while the stability is ensured by the multi-stage buffer region protection with the sound quality priority; a multi-device coordination and timing monitoring mechanism ensures perfect cooperation with a third-party device, and the stability challenge in high-resolution audio transmission is comprehensively solved.

The invention relates to the technical field of sound quality enhancement, in particular to a sound quality improvement method and device of a Bluetooth loudspeaker box, equipment and a storage medium. The method comprises the following steps: collecting sound box environment acoustic signals, and carrying out peripheral audio time-frequency spectrum analysis and three-dimensional sound field fitting to construct a peripheral three-dimensional sound field model; based on the surrounding three-dimensional sound field model, environment noise source semantic deep analysis is carried out, and user real-time scene sound quality demand prediction is carried out, so that real-time scene sound quality demand features are generated; bluetooth sound box loudspeaker monitoring parameters are collected, and adaptive noise frequency suppression adjustment is carried out based on the surrounding three-dimensional sound field model rate, so that a loudspeaker dynamic noise suppression adjustment strategy is generated; and acquiring a Bluetooth audio output signal, and performing band frequency response characteristic optimization according to the real-time scene tone quality demand characteristics to obtain a frequency response tone quality optimized audio signal. According to changes of different environments and requirements, the output tone quality of the Bluetooth loudspeaker box is optimized in real time, and high-fidelity audio output is kept.

The invention relates to the technical field of earphone noise reduction processing, and discloses a motorcycle riding earphone noise reduction method combining different noise reduction algorithms, comprising the following steps: acquiring an original audio signal, and distributing the original audio signal to each processing module; a preprocessing signal is generated, voice activity information is determined, and environmental noise energy information is calculated; according to the environmental noise energy or the auxiliary information, adaptively adjusting a high-low noise energy threshold value; the original signals are input into the traditional and AI noise reduction module in parallel to generate two paths of noise reduction signals; based on the environment noise, the voice information and the threshold value, performing weighted fusion to generate an output signal; and carrying out equalization and compression processing on the output signal, and driving the loudspeaker to output. According to the invention, comprehensive judgment on environmental noise energy information, voice activity information and riding speed is introduced, an output signal of a traditional noise reduction module is set to be zero in a high-speed voice-free environment, and AI noise reduction is combined, so that tone quality distortion of a traditional noise reduction algorithm in a noise environment is effectively avoided.

The invention relates to the technical field of audio coding and decoding, and discloses an audio coding and decoding method and system, and the method comprises the steps: carrying out the time domain analysis and frequency domain analysis of an audio input signal of a Bluetooth earphone, and generating an audio feature matrix; performing feature extraction and fusion processing to generate a target feature vector containing frequency band energy distribution and user preference information; inputting the target feature vector into a preset neural network model for dynamic parameter calculation, and generating an adaptive coding matrix; performing layered division on the audio input signal according to the adaptive coding matrix to obtain basic tone quality layer data and enhanced tone quality layer data; and encoding the basic tone quality layer data and the enhanced tone quality layer data through a heterogeneous computing unit to generate a target encoded data stream, and the method realizes double optimization of audio processing precision and processing time.

The invention discloses an audio data compression method and device, electronic equipment and a storage medium, relates to the technical field of voice processing, can be applied to financial science and technology and medical health business scenarios, and comprises the following steps: obtaining target audio data to be compressed; and inputting the target audio data into an audio coding and decoding quantization compression model to obtain a reconstructed audio signal corresponding to the target audio data, the audio coding and decoding quantization compression model being obtained through audio adversarial training. In the model, feature extraction and compression can be performed on target audio data by using a hierarchical neural network of an encoder to obtain a low-dimensional potential feature vector; a residual vector quantization module performs discrete quantization processing on the low-dimensional potential feature vector through a multi-layer cascaded codebook to obtain a discrete quantization code; and performing audio waveform reduction processing on the discrete quantization code by using a decoder to obtain a reconstructed audio signal corresponding to the target audio data. According to the invention, audio high-fidelity compression can be realized, and audio reconstruction tone quality is improved.

The invention provides a self-adaptive adjustment method for sound tone of a live broadcast person based on a network model, which relates to the technical field of audio signal processing and comprises the following steps of: determining an accurate starting position and an accurate ending position of packet loss through a sudden change point of a short-time energy function and an abnormal change point of a zero-crossing rate; therefore, the packet loss interval is repaired by pertinently utilizing the fundamental frequency track characteristic and the formant characteristic. The method provided by the invention can solve the problems of fundamental frequency track breakage and unnatural voice caused by the contradiction between acoustic parameter decoupling repair and real-time processing constraint in a high packet loss rate scene in a traditional voice repair method, thereby realizing high-precision and low-delay self-adaptive repair of packet loss voice, and improving the voice recovery efficiency. And the sound quality and the user experience in the live scene are obviously improved.

The invention relates to a self-adaptive frequency optimization system in radio and television microwave transmission, and the system comprises an initial audio collection and processing module which collects an audio signal of a current frequency channel in real time, and generates a first data set; the conversion audio acquisition processing module synchronously acquires a target channel signal according to the frequency switching signal, and calculates the loudness offset between two channels; a feature extraction module extracts frequency domain energy distribution feature vectors of the first data set and the second data set; the model construction module constructs a dynamic gain compensation model and generates a compensation value of a target channel; the tone quality compensation module performs smoothing processing on the compensation value to generate an optimized third data set; and the audio optimization module updates model parameters, dynamically adjusts a frequency domain analysis strategy, and outputs low-delay and high-stability radio and television transmission signals. According to the invention, the problems of high switching delay and degraded tone quality caused by static configuration in a traditional scheme are solved, and the transmission efficiency and user experience in a complex electromagnetic environment are remarkably improved through modular dynamic calibration and compensation.

The invention relates to the technical field of audio processing, in particular to a self-adaptive scene sound effect adjusting method and device and a storage medium, and the method comprises the steps: collecting a mixed audio stream and an original playing audio, and extracting an acoustic fingerprint vector through a pre-trained convolutional neural network; performing acoustic feature deconstruction on the acoustic fingerprint vector to obtain a plurality of acoustic feature components, performing scene recognition judgment according to a preset scene judgment rule based on the plurality of acoustic feature components, and determining an audio playing scene; constructing a time-frequency mask according to the audio playing scene, generating a noise reduction gain matrix based on the time-frequency mask, and performing scene sound field enhancement on the audio playing scene to obtain a sound field optimization matrix; and performing noise reduction and sound effect adjustment on the original playing audio according to the noise reduction gain matrix and the sound field optimization matrix, and outputting the optimized playing audio. According to the invention, sound effect adjustment can be dynamically adapted according to different scenes, and the noise reduction accuracy and the multi-scene sound quality adaptability are improved.

The invention provides a sound field optimization method and device of a multichannel loudspeaker array, equipment and a medium, and relates to the technical field of acoustics, to-be-played audio signals are respectively input to each phase compensation filter for phase compensation filtering, and a plurality of phase-compensated audio signals are obtained; the phase compensation filter is determined according to the phase compensation value of the loudspeaker; respectively inputting each phase-compensated audio signal into a corresponding loudspeaker so as to control the multi-channel loudspeaker array to produce sound; acquiring sound field response data in the target area; and when the difference between the sound field response data and the target sound field data does not meet the set sound effect requirement, dynamically adjusting the phase compensation value of the loudspeaker, and repeating the step of phase compensation filtering until the difference between the obtained sound field response data and the target sound field data meets the set sound effect requirement. According to the invention, through a dynamic adjustment mechanism and a closed-loop optimization process, precise regulation and control of the sound field of the loudspeaker array are realized, and the sound effect quality in the target area is significantly improved.

Methods and apparatus are described herein related to improving the sound quality of a bone conduction speaker. The sound quality of the bone conduction speaker is adjusted in the sound generation, sound transferring, and sound receiving of the bone conduction speaker by designing vibration generation manners and vibration transfer structures.

The invention discloses a meeting window call system based on a voice sensor, and relates to the technical field of communication. Comprising a full-space voice acquisition module, a dynamic confidence evaluation module, a phase coupling enhancement module, a multi-dimensional speech mode recognition module, a dynamic gain adjustment and tone quality enhancement module and a full-link voice integrity verification module, an array type voice sensing unit is deployed based on audio reflection characteristics, a multi-band full-space acquisition matrix is constructed, voice signals of two parties meeting are acquired, and a basic audio data stream with a high signal-to-noise ratio is output. According to the invention, through array sensing acquisition, dynamic confidence evaluation, phase coupling and harmonic enhancement, multi-dimensional speech recognition and dynamic gain adjustment, accurate recognition and fidelity enhancement of low-volume speech are realized, and through combination of full-link integrity verification and mistaken killing backtracking, weak signals are ensured not to be missed and distorted, so that the accuracy of speech recognition is improved. And the privacy, the security and the information integrity of the meeting call are improved.

The invention discloses a call recording method based on dynamic sampling rate adjustment and a Bluetooth earphone, and aims to improve the quality and stability of call recording in a wireless transmission environment. According to the method, uplink audio data and downlink audio data are collected at the same time through the Bluetooth headset, and the transmission rate of a downlink data packet is monitored in real time. When the rate is lower than the preset threshold value, the audio sampling rate is automatically reduced from 16 kHz to 8 kHz to adapt to bandwidth limitation, and recording interruption or tone quality reduction is avoided. The adjusted audio data is cached through a double-layer cache mechanism after being coded and compressed, the first cache temporarily stores the adjusted audio data, and the second cache manages data packets to be sent and retransmitted, so that the continuity and integrity of data transmission are ensured. The method is suitable for wireless audio equipment such as a Bluetooth earphone, and the reliability, adaptability and user experience of call recording are effectively improved.

A loudspeaker assembly and a headphone are provided. The headphone includes a housing assembly, a bone-conduction core module, and an air-conduction core module, and an accommodation space. The bone-conduction core module is provided in the accommodation space and vibrates in a first vibration direction, the air-conduction core module is provided in the accommodation space, and the bone-conduction core module and the air-conduction core module are arranged along the first vibration direction and face each other. In the above manner, the present disclosure reduces the influence of the air-conduction core module on the vibration effect of the bone-conduction core module, increases the sound volume, and improves the bone-conduction effect of the loudspeaker assembly, thereby improving the sound quality effect of the loudspeaker assembly.

The invention relates to the cross technical field of artificial intelligence and multimedia communication, and discloses an intelligent optimization system and method fusing video resource scheduling and voice emotion recognition, and the system comprises a voice emotion recognition module which is used for collecting a user voice stream in real time, extracting acoustic features through a deep learning model, carrying out the emotion classification, and obtaining a voice emotion recognition result; and outputting the emotion label. According to the invention, through voice emotion recognition, the system realizes personalized resource allocation and content adjustment, so that the personalization and intelligence of user experience are greatly improved; the video resource scheduling module dynamically adjusts parameters according to a network, equipment and emotions to ensure smooth playing of a video and excellent image quality and tone quality; a multi-modal decision engine fuses a multivariate data generation strategy, resource utilization is optimized, and the emotion matching degree and the QoS score are improved; the intelligent optimization method is clear in process, video resources are accurately optimized, strategies are continuously updated through user feedback, the system adaptability is enhanced, and high-quality and personalized multimedia service experience is comprehensively provided for users.

The invention discloses a debugging method and system of an audio system, and relates to the technical field of audio debugging, the method comprises the steps of S1, environmental acoustic analysis, S2, basic parameter presetting, S3, real-time monitoring and dynamic adjustment, and S4, adaptive optimizing.The layout of an equalizer or a loudspeaker is adjusted through environmental acoustic analysis; presetting basic parameters of the sound amplifying device and the microphone; real-time multi-band sampling, delay calibration and gain regulation are realized; adaptive optimization is carried out, algorithms such as noise reduction are started, stored parameters are scene presetting, and one-key calling and self-checking updating are supported; the environment is analyzed, the audio system is accurately debugged, the system is dynamically optimized, the tone quality and the debugging efficiency are improved, and diversified scene requirements are met.

Methods and apparatus are described herein related to improving the sound quality of a bone conduction speaker. The sound quality of the bone conduction speaker is adjusted in the sound generation, sound transferring, and sound receiving of the bone conduction speaker by designing vibration generation manners and vibration transfer structures.

The invention relates to an electrostatic vibrating diaphragm automobile audio which is mainly used for an electric vehicle and can compensate vehicle-mounted electromagnetic interference audio and solve the problem of electromagnetic interference of a motor PWM switch and rotating speed harmonics on an audio system. The electrostatic diaphragm automobile sound system comprises an electrostatic diaphragm loudspeaker and an interference feature extraction module, the rotating speed of a motor and the PWM frequency are obtained through a CAN bus, and a dynamic interference feature set is analyzed and extracted through adaptive window length FFT; the EMI quantification module is used for calculating an electromagnetic pollution index at a frequency band of 8kHz-20kHz, and triggering compensation when the electromagnetic pollution index exceeds a threshold value of-40dB; the dynamic notch compensation module is used for generating an inversion signal; according to the double-loop feedback control module, the amplitude is adjusted through PID of an inner loop, and the phase is optimized through gradient descent of an outer loop so as to minimize THD. The method comprises five steps of signal acquisition, feature extraction, interference quantification, dynamic compensation and signal injection. Interference frequency drift can be accurately tracked, a tone quality MOS value is improved after compensation, and an over-compensation risk is avoided.

The invention belongs to the technical field of loudspeakers, and discloses a bidirectional loudspeaker group, a sound box and audio equipment, and the bidirectional loudspeaker group comprises a common magnetic circuit module, a first loudspeaker and a second loudspeaker. The common magnetic circuit module comprises a magnetic conduction bowl, a first magnetic gap and a second magnetic gap are formed in the inner side and the outer side of the magnetic conduction bowl in the first direction respectively, a first through hole communicated with the first magnetic gap and the second magnetic gap is formed in the magnetic conduction bowl, and the first loudspeaker and the second loudspeaker are arranged on the two sides of the common magnetic circuit module in the second direction respectively. The first loudspeaker comprises an annular first voice coil, the second loudspeaker comprises an annular second voice coil, the first voice coil and the first magnetic gap are correspondingly arranged, the first voice coil can vibrate in the first magnetic gap in the second direction, the second voice coil and the second magnetic gap are correspondingly arranged, and the second voice coil can vibrate in the second magnetic gap in the second direction. Through the above arrangement, the bidirectional loudspeaker group provided by the invention can achieve the balanced damping, sound quality improvement and heat dissipation requirements.

The present application provides an audio processing method and apparatus, a wireless earphone, and a storage medium. A first wireless earphone receives a first to-be-presented audio signal sent by a playing device, and a second wireless earphone receives a second to-be-presented audio signal sent by the playing device; then, the first wireless earphone performs rendering processing on the first to-be-presented audio signal to obtain a first audio playing signal, and the second wireless earphone performs rendering processing on the second to-be-presented audio signal to obtain a second audio playing signal; and finally the first wireless earphone plays the first audio playing signal, and the second wireless earphone plays the second audio playing signal. Therefore, it is possible to achieve technical effects of greatly reducing the delay and improving the sound quality of the earphone since the wireless earphone can render the audio signals independently of the playing device.

The invention discloses a multi-channel audio calibration method, system and device and a storage medium, and relates to the technical field of audio, and the method comprises the steps: collecting and separating a main sound channel signal and a bass sound channel signal in a multi-channel audio signal; detecting time delay between a main sound channel and a bass sound channel based on a phase correlation algorithm, and performing adaptive delay compensation; the state change of the audio signal is monitored in real time, and before or when a signal switching event is detected, gradient gain envelope is applied to the bass sound channel signal, so that the signal amplitude gradually changes to a target value according to a preset nonlinear curve; a sonic boom event is analyzed and identified based on multi-dimensional characteristics, a self-adaptive amplitude limiter with a prospective processing capability is started for suppression, and harmonic distortion compensation and phase correction are carried out on a processed signal, so that the problems of signal asynchronization caused by low-frequency output delay of an audio system in the prior art, and the system reliability is improved are solved. Therefore, the low-frequency sound quality and the system stability are improved.

The invention discloses a subjective and objective fused air conditioner sound quality analysis and evaluation method, which comprises the following steps of: 1, forming a working condition-gear combination matrix according to an actual use scene and noise characteristics of an air conditioner; 2, arranging sensor measuring points according to a noise source scene of working condition-gear combination, and obtaining a sound sample; 3, carrying out subjective evaluation on the sound sample, and processing an abnormal score by adopting a 3 sigma principle and a box plot method to obtain a final subjective score; 4, five objective psychological acoustic indexes including loudness, roughness, sharpness, jitter degree and highlight ratio of 24 Bark frequency bands of the sound sample are calculated, and an objective index set is established; and 5, constructing an air conditioner subjective and objective fusion evaluation model of a two-stage fusion model of the time sequence convolutional network and the gradient boosting tree. According to the method, the problem of low evaluation precision of a general model caused by noise characteristic difference under different working conditions-gears is solved, an accurate air conditioner sound quality evaluation system is constructed, and support is provided for sound quality optimization design and quality evaluation of the air conditioner.

The invention relates to the technical field of voice processing, in particular to a sound source direction-finding positioning method and system for a complex environment, and the method comprises the steps: serializing the sound intensity of each microphone, and calculating a maximum receiving interval according to the distance between adjacent microphones, the sound transmission speed and the sampling frequency; the method comprises the following steps: determining a short-time audio similarity through moving a sound source signal sequence, extracting a sound source synchronization sequence and a sound source same-frequency sequence, and further evaluating a tone quality deviation degree; optionally selecting a reference microphone, and comparing the reference microphone with a sound source synchronization sequence of other microphones to obtain a time delay data length and a sound source similar sequence; based on the arrangement values of the same position elements of the sound source similar sequence, a sound intensity distribution sequence is obtained, and the sound intensity interference degree of each microphone is obtained; and determining a fixed beam direction and calculating the distance between the reference microphone and the sound source. The invention aims to improve the accuracy of sound source positioning.

The invention provides an acoustic system control method based on a scene type and a related device, and the method comprises the steps: carrying out the comprehensive analysis of a scene based on the collected real-time scene data in a current site, and obtaining the scene type of the current site at each time; when it is monitored that the current site is switched from the first scene type to the second scene type, based on a first acoustic parameter corresponding to the first scene type and a second acoustic parameter corresponding to the second scene type, an acoustic system is controlled to be transitionally switched to a target acoustic parameter; environment data of the current site at the time corresponding to the second scene type are collected; optimizing the second acoustic parameter based on the environmental data to obtain an optimized acoustic parameter; and controlling acoustic output parameters of an acoustic system of the current site based on the optimized acoustic parameters. According to the method, the problem of adapting to dynamic changes of field acoustic characteristics is solved, a good acoustic effect can be obtained in various scenes and environments, and the sound quality of an acoustic system is improved.

A metal-free signal cable made of at least 24 carbon rovings twisted together for use in music reproduction as a loudspeaker cable, small signal cable or bridge, which is kept free from external influences by spacers made of electromagnetically practically neutral aluminum-magnesium alloy. Even the highest-quality hi-fi systems sound significantly more energetic, clearer, more accurate, more tangible, and even more "in the room" when using my invention. Transients are much more realistic, the reproduced stage size increases significantly, the sound and positioning of the instruments involved are noticeably improved, and each individual note becomes significantly more subtle, complex, and rich in nuance. Even the smallest sound events become perceptible, which are lost with conventional solutions.

The invention relates to a sound box built-in power amplifier remote audio transmission and control method based on 5G communication, and the method comprises the steps: obtaining real-time monitoring data in an audio data transmission process, and generating an initial delay distribution data set for each audio channel; analyzing network delay fluctuation of each audio channel through a timestamp calibration method, and determining a delay reference value; setting a delay deviation threshold range according to the delay reference value, identifying an abnormal channel and recording a delay offset; analyzing a phase relation of the audio data stream after time correction, calculating a phase deviation angle and determining a phase adjustment parameter; performing phase correction on the audio data stream through a digital signal processing module to generate audio signals with consistent phases; detecting tone quality consistency difference among multiple devices, adjusting audio output feature data and optimizing a time window configuration scheme; and dynamically updating the audio channel coordination strategy, and storing correction parameters to a system database.

The invention discloses a directional pickup method for a microphone. The microphone comprises a left microphone core, a middle microphone core and a right microphone core. Wherein the middle microphone core is located on the central axis of the microphone, and the left microphone core and the right microphone core are symmetrically distributed on the left side and the right side of the middle microphone core; sound of a sound source is synchronously collected through a left microphone core, a middle microphone core and a right microphone core of the microphone, and three paths of sound signals are obtained; respectively filtering the three paths of sound signals to obtain left, middle and right microphone core signals; calculating an included angle between a connecting line of a sound source and the middle microphone core and the central axis of the microphone based on the time difference and the distance difference of the sound reaching the left, middle and right microphone cores to obtain a sound source direction; based on the left microphone core signal, the middle microphone core signal, the right microphone core signal and the sound source direction, through time delay compensation and gain adjustment, the sound signal in the target direction is enhanced, the sound signal in the non-target direction is suppressed, the final output signal is obtained, and the directionality and the sound quality of pickup are improved.

The present disclosure provides an audio noise reduction method, apparatus, device, and storage medium. The method includes: First, obtaining audio data to be noise-reduced, and then using a preset real number network model to estimate the magnitude time-frequency mask of the audio data to be noise-reduced, so as to obtain a first-order enhanced magnitude spectrum corresponding to the audio data to be noise-reduced. Furthermore, using a preset complex number network model to estimate the complex time-frequency mask of the audio data to be noise-reduced, and combining the first-order enhanced magnitude spectrum and the complex time-frequency mask to determine the noise-reduced result audio data corresponding to the audio data to be noise-reduced. In the embodiments of the present disclosure, the magnitude spectrum of the audio data to be noise-reduced is enhanced by using a preset real number network model, and the magnitude spectrum and phase spectrum of the audio data to be noise-reduced are simultaneously enhanced by using a preset complex number network model. It can be seen that the embodiments of the present disclosure can implement noise reduction processing on the audio data to be noise-reduced, thereby better improving the audio quality.

The invention discloses an AI-based high-fidelity audio coding and decoding algorithm adaptive to an auditory range and a compression optimization method, and the algorithm comprises an audio signal preprocessing module which is used for carrying out the preprocessing of an input audio signal, including but not limited to denoising, frequency filtering, gain adjustment, time domain / frequency domain transformation and the like, and carrying out the compression optimization of the audio signal; and the preprocessing module is used for performing feature extraction through a convolutional neural network (CNN) or an auto-encoder network so as to improve the signal-to-noise ratio of an audio signal and remove environmental noise, and generating optimized audio features. Through an AI-based auditory range adaptation algorithm, dynamic optimization can be performed according to individual auditory characteristics, so that the audio coding and decoding process is more in line with the perception characteristics of human ears, better balance is obtained between the compression ratio and the tone quality, higher audio experience is provided, and through the adaptive quantization and code rate control technology, the audio coding and decoding efficiency is improved. According to the method, on the premise of keeping the high-fidelity tone quality, the deep learning algorithm is utilized to optimize the allocation of computing resources, and the consumption of the computing resources is reduced.