151 results about "Dynamic noise" patented technology

The invention relates to the technical field of wind turbine generator state monitoring, in particular to a wind turbine generator state monitoring method and system based on multi-source heterogeneous data fusion, and the system comprises a data collection unit, a dynamic noise processing unit, a damage feature analysis unit and a health state output unit. A data acquisition unit synchronously obtains blade strain, unit rotating speed and environment vibration signals through hardware timestamp alignment, and a dynamic noise processing unit removes rotating speed coupling noise by using a two-parameter coupling model, a temperature-pitch angle three-dimensional correction curved surface and closed-loop feedback in combination with variable step blanking and local band elimination protection. The damage feature analysis unit extracts microcrack features by adopting multi-scale wavelet packet decomposition and kurtosis detection, and performs cross validation by fusing a vibration mode confidence factor, and the health state output unit generates a topological graph containing a crack position, an expansion trend and an alarm confidence level, so that the problems of insufficient multi-source data fusion and false and missing judgment of damage are solved, and the safety of the system is improved. And the monitoring precision is improved.

The invention discloses an operation AR intelligent auxiliary management system based on voice AI interaction driving, which is applied to the field of intelligent warehousing and comprises an environment perception type voice interaction module, an AR multi-mode interaction compensation module, a dynamic noise robustness data processing module, an operation management module and a database module. The environment perception type voice interaction module is used for collecting voice signals of operating personnel and inhibiting environment noise; the dynamic noise robustness data processing module is used for carrying out noise reduction processing, semantic understanding and closed-loop optimization on the voice signal and generating feedback information and a control instruction; the AR multi-modal interaction compensation module is used for receiving feedback information of the dynamic noise robustness data processing module and providing visual and tactile interaction compensation, and the job management module is used for scheduling management equipment according to a control instruction; according to the method, the operation efficiency can be effectively improved, the accuracy is high, the real-time performance is high, and interaction experience optimization can be realized.

The invention relates to the field of adaptive noise reduction, in particular to an adaptive noise reduction method and system for a multi-mode audio SoC main control chip. The method comprises the following steps: acquiring an original audio input signal according to an SoC main control chip, performing time-frequency domain dual deconstruction and cross-frequency domain noise interference structure analysis, and constructing a full-frequency domain noise interference topology table; carrying out multi-mode interference factor separation on the full-frequency-domain noise interference topology table, carrying out multi-noise environment modeling, and constructing a real-time noise scene model; time sequence noise slope fluctuation modeling is carried out on the real-time noise scene model, dynamic noise reduction response learning optimization is carried out, and a dynamic noise reduction optimization strategy is constructed; and performing neural coding gain and audio boundary line reconstruction according to the original audio input signal to obtain a coding gain key audio signal and a weak audio optimization signal. According to the invention, by flexibly adjusting the noise reduction intensity of the audio signal, the real-time scene noise reduction performance is optimized.

The invention relates to the technical field of simultaneous interpretation, and discloses a simultaneous interpretation data processing method and system based on a POE microphone array. The method comprises the following steps: synchronously acquiring multi-language original audio streams and meeting place environment noise spectrum features through a distributed microphone array powered by the Ethernet; after time domain framing is carried out on the audio stream, adaptive filtering is carried out by using a dynamic noise reduction weight coefficient to obtain a primary pure voice segment; dividing the multi-language speech endpoint detection model into independent speech units with language labels through a pre-trained multi-language speech endpoint detection model, and matching a corresponding acoustic model to generate a phoneme-level time alignment sequence; comparing and outputting a term replacement instruction stream in real time in combination with a simultaneous transfer term library, and generating an intermediate semantic representation vector after fusion; and the low-delay encoder converts the voice parameter sequence into a target language voice parameter sequence, and drives the waveform synthesizer to generate final simultaneous transmission audio. The method optimizes the whole process processing, gives consideration to the simultaneous transmission accuracy and real-time performance, and is suitable for a multilingual meeting place scene.

The invention discloses a self-adaptive filtering method and system for an intelligent electric energy meter, and relates to the technical field of power grid signal processing, and the method comprises the steps: obtaining a power grid voltage and current signal set collected by a target intelligent electric energy meter historically, and obtaining a signal noise spectrum data set and a noise level evaluation result through spectrum analysis and noise level evaluation; if the result exceeds a preset noise threshold value, identifying an interference type and building a noise adaptive filtering multi-channel, integrating and deploying the noise adaptive filtering multi-channel into a target intelligent electric meter to complete signal filtering test and effect evaluation, and dynamically optimizing and generating an electric energy meter adaptive filtering multi-channel based on the obtained multi-channel filtering effect index parameter; and finally, signal filtering closed-loop control is realized. The technical problem that a traditional electric energy meter is weak in filtering pertinence and cannot adapt to dynamic noise of a power grid in electric power related scenes with complex dynamic interference is solved, and the technical effects that the electric variable measurement accuracy of the electric energy meter is improved, and the adaptive capacity of the electric energy meter to different interference scenes is enhanced are achieved.

The invention discloses a voice data recognition method and system based on an AI voice algorithm, relates to the technical field of AI voice recognition, and solves the problem that the voice data recognition capability is low. The method comprises the following steps: S1, multi-mode cooperative triggering collection: synchronously collecting lip electromyographic signals and voiceprint features through a multi-mode sensor, an activation instruction is generated through feature fusion, and voice acquisition starting is triggered; s2, AI adaptive noise reduction processing: carrying out noise separation on the original audio signal by adopting a generative adversarial network, separating environmental noise features to generate a dynamic noise reduction mask, and keeping the integrity of human voice features; s3, beam dynamic optimization adjustment: analyzing real-time audio quality based on a reinforcement learning algorithm, dynamically adjusting beam pointing and gain parameters of a microphone array, and focusing a target sound source; and S4, semantic association cache enhancement: carrying out real-time semantic analysis on the collected voice data. According to the invention, the voice data recognition capability of an AI voice algorithm is greatly improved.

The invention discloses a real-time noise reduction method and system supporting Bluetooth audio interaction, and the method comprises the steps: collecting an environment audio signal and a Bluetooth interaction audio signal, and carrying out the timestamp alignment and spatial calibration of the environment audio signal and the Bluetooth interaction audio signal; an improved multivariate variational mode decomposition algorithm is adopted to decompose and extract noise features, the noise features are combined with user historical noise data, a dynamic noise model is constructed through a long-short-term memory network, parameters are updated in real time, Bluetooth interaction audio content features and a user behavior data recognition scene are analyzed, and a noise reduction strategy is determined. And according to the dynamic noise model and the scene result, reinforcing learning to optimize filter parameters, generating a directional anti-noise signal to carry out noise reduction on the Bluetooth interaction audio signal, collecting the noise reduction intensity manually adjusted by a user, calculating voice definition and total harmonic distortion, and carrying out noise reduction on the Bluetooth interaction audio signal. And inputting a feedback and evaluation result into the modeling and noise reduction steps, carrying out dynamic range adjustment, sound channel balance and power amplification on the audio after noise reduction, and outputting the audio through a loudspeaker.

The invention relates to a differential privacy position protection method based on a dynamic attenuation noise field, and the method comprises the steps: obtaining a plurality of candidate regions based on a user track, recognizing different sensitive regions from the plurality of candidate regions based on the behavior characteristics of a user, and calculating the dynamic sensitivity of each sensitive region; wherein the behavior characteristics of the user comprise a stay identifier and access time of the user in the candidate area; constructing a three-dimensional dynamic noise model for generating a dynamically attenuated space-time noise field; the method comprises the following steps: acquiring a position coordinate of a target position in a sensitive area and a corresponding timestamp, generating a corresponding noise amplitude by adopting a three-dimensional dynamic noise model, generating a two-dimensional position offset based on the noise amplitude by adopting a Laplace mechanism, and disturbing the position coordinate of the target position based on the two-dimensional position offset. According to the scheme, adaptive adjustment of the noise amplitude can be realized, and collaborative optimization of strong protection of a high-sensitivity area and high precision of a low-sensitivity area is realized.

The invention relates to the technical field of sensing, and discloses a capacitive touch screen input event classification and processing system which comprises an input capture module, an event classification module, an intention prediction module, a dynamic response module, an energy efficiency optimization module and a multi-channel cooperation module. When an input event of the capacitive touch screen is identified, a dynamic noise suppression model is established, and adaptive filtering strategies are configured for different use scenes, so that the reliability of signal analysis in various electromagnetic interference environments is ensured, and meanwhile, the spatial and temporal distribution characteristics of capacitance variation are analyzed in real time; real touch operation and environmental electromagnetic noise can be effectively distinguished, the identification fidelity of basic operations such as clicking and sliding is ensured, the false triggering rate is reduced, the operation accuracy in a weak signal scene is improved, and when multi-finger operation is detected, the consistency of contact coordinates in a curved surface state and a plane state is maintained, so that the accuracy of the touch operation is improved. And the track continuity of writing and drawing type fine operation is ensured.

The invention provides a battery thermal runaway early warning method and device. The method comprises the following steps: acquiring an original acoustic signal released by a target battery pack by using an acoustic sensor; working condition information of the target battery pack is obtained, dynamic noise reduction processing is performed on the original acoustic signal based on the working condition information, multi-source interference noise is suppressed, a de-noised optical signal is obtained, and acoustic characteristics of the de-noised optical signal in a preset frequency band are extracted; constructing a thermal runaway acoustic triggering index based on the energy threshold value, the combined characteristic change threshold value and the continuous triggering time, and generating an initial early warning signal according to the thermal runaway acoustic triggering index and the acoustic characteristics; state information of the target battery pack is obtained, multi-mode fusion analysis is carried out based on the state information and the initial early warning signal, and a thermal runaway verification signal is generated; and based on the thermal runaway verification signal and the initial early warning signal, performing thermal runaway grading early warning on the target battery pack. According to the invention, the accuracy of thermal runaway early warning can be improved.

The invention discloses a household electrical appliance fault diagnosis system based on artificial intelligence, and relates to the technical field of household electrical appliance fault diagnosis and state monitoring. Working condition parameters aligned with a time window are obtained through working condition sensing, and fault features and the working condition parameters are linked in an order form to be converted into a target frequency track; fault characterization is converted into track characterization capable of being continuously tracked along with working conditions from a fixed frequency point, and disturbance to fault positioning in the speed change process is reduced; in order to solve the problem that weak fault components are covered by the same frequency band due to strong dynamic background noise jointly formed by inverter switch harmonic waves, current harmonic waves and load fluctuation, a dynamic noise reference spectrum is output under the current working condition through a condition generation model, and a current spectrum and the dynamic noise reference spectrum form a residual spectrum; the background spectrum base related to the working condition is suppressed, and the separability of the weak fault component in the residual domain is improved from the source.

The invention discloses a real-time processing method based on voice short messages, and relates to the technical field of communication, and the method comprises the specific steps of voice preprocessing, voice transcription and text specification, semantic analysis and processing, information processing and result feedback. According to the invention, through an adaptive filtering formula in voice preprocessing and a segmentation mechanism based on voice pause intervals and intonation changes, dynamic noise reduction processing and accurate segmentation of voice signals are realized, and the frequency and amplitude of noise signals and corresponding characteristics of the voice signals can be analyzed in real time. A filtering coefficient is dynamically adjusted through a self-adaptive filtering formula, non-voice signal components are effectively filtered out, meanwhile, complete voice information is reserved, continuous voice streams are accurately segmented by capturing pause intervals and intonation changes in voice signals, standardized voice segments with timestamps and sequence identifiers are formed, and therefore the voice processing efficiency is improved. And a high-quality data basis is provided for subsequent voice transcription and semantic analysis.

The invention discloses an audio signal measurement method and device, an electronic device and a storage medium, and relates to the field of audio signal measurement, and the method comprises the steps: deploying a detection microphone nearby a tested device, and arranging a reference microphone in a noise field, so as to synchronously collect a mixed signal containing a test signal and noise, and a pure ambient noise reference signal. After the noise of the two signals is subjected to correlation evaluation and the consistency of noise sources is confirmed, dynamically changing noise components are accurately stripped from the signals of the tested equipment by utilizing real-time noise data acquired by the reference microphone and through an adaptive noise cancellation algorithm, so that pure test signals are restored. Therefore, the problem of non-stationary noise pollution is solved through an active signal processing means instead of depending on imperfect physical sound insulation, and accurate and efficient measurement of the performance of the microphone in a noisy environment is realized.

The invention relates to the technical field of gas concentration detection, in particular to a gas concentration detection method and system for a ring main unit. The method comprises the following steps: acquiring a concentration time sequence of a gas sensor in the ring main unit; constructing an instantaneous fluctuation coefficient; analyzing the instantaneous fluctuation coefficient and a gas physical diffusion limit rate constant to obtain an environmental interference factor representing electromagnetic interference intensity; obtaining an environment interference curve, and combining the environment interference factor and the integral operation value of the environment interference curve to obtain a dynamic adjustment parameter; a dynamic noise covariance used for Kalman filtering is obtained; and performing Kalman filtering processing on the concentration time sequence by using the dynamic noise covariance to obtain an optimal estimated gas concentration value. The problems of false alarm and missing detection in the gas concentration detection process are solved.

The invention relates to the technical field of intelligent communication, in particular to an AI (Artificial Intelligence)-based communication voice noise reduction method, which comprises the following steps of: 1, collecting noise of multiple scenes, recording noisy voice through a microphone array, and covering complex environments such as traffic, restaurants, meetings and the like; 2, constructing a dialect voice database, adding a noise mixing layer (such as sudden whistling and keyboard tapping) and a scene evolution layer, and simulating a real environment; according to the method, multi-scene noise is collected, noisy voice is recorded through a microphone array, a dialect voice database is constructed, a noise mixing layer is added, voice features of different dialects in a complex environment are simulated, a deep learning model is deployed on edge equipment, and low-delay noise reduction is achieved through real-time processing. Mechanical noise, aerodynamic noise and external environment noise are distinguished through the dynamic noise recognition module, the noise reduction intensity is adjusted to adapt to scenes such as elevators, meeting rooms and vehicles, and therefore the method can adjust the noise reduction intensity in a self-adaptive mode.

The invention belongs to the field of aircraft digital assembly automatic drilling and riveting, and particularly discloses an automatic drilling and riveting machine actuator normal alignment method and system.The method comprises the steps that firstly, the original distance from the end face of a presser foot to the plane around a to-be-machined hole is collected through a sensor; then, a KF-MAF algorithm combining Kalman filtering and moving average filtering is adopted to suppress dynamic noise, and accurate distance information is obtained; then, the deviation between the feeding direction of an actuator and the normal direction of the surface of the hole to be machined is calculated, and the angle error amount of a target hole site is obtained through a bilinear interpolation method; and finally, an actuator is controlled to rotate to compensate errors, the position of the center point of the tool is kept unchanged, the axis of the drill bit coincides with the local surface of the workpiece in the normal direction, and normal alignment is achieved. According to the method, the needed original information is less and easy to obtain, alignment can be accurately and rapidly completed, and the drilling precision is guaranteed.

The invention discloses a building engineering environment monitoring method and system, and particularly relates to the technical field of environment monitoring, and the method comprises the steps: S1, building an initial monitoring network topology structure; s2, generating composite noise source recognition features; s3, generating a node operation load degree; s4, monitoring a priority adjustment instruction; and S5, integrating the data to form a hierarchical environment state data set. According to the construction engineering environment monitoring method and system, the specific structure size of the tower crane is obtained through the laser range finder, and the initial layout of the monitoring points is realized in combination with the sound wave attenuation model, so that the traditional gridding or empirical point distribution is abandoned, and the construction engineering environment monitoring efficiency is improved. According to the method, the monitoring network has the optimized coverage capability aiming at a core dynamic noise source from the beginning of construction, the monitoring signal-to-noise ratio of a key area is improved, and in the noise identification stage, a sliding time window cross-correlation algorithm is adopted to carry out synchronous analysis on two specific noise sources of the tower crane and the pump truck.

The invention provides an intelligent audio power amplifier system and a working method thereof. The working method comprises the steps that a dynamic noise suppression module enables an input signal after reverse noise injection to enter a distortion compensation module; a distortion compensation signal is synthesized by using a distortion compensation module; the life cycle parameter self-calibration module calibrates parameters and inputs the parameters to the dynamic noise suppression module and the distortion compensation module; the dynamic noise suppression module is based on an intelligent learning method and adopts a frequency band channel network, a time sequence channel network and a fusion layer; the life cycle parameter self-calibration module provides the temperature and the working time for the dynamic noise suppression module to optimize the time domain channel network and provides the temperature and the working time for the distortion compensation module to update the predistortion coefficient of the dynamic nonlinear distortion model; the output of the dynamic noise suppression module is provided for the distortion compensation module; and the output of the distortion compensation module is fed back to the dynamic noise suppression module. The problems of high-frequency noise interference, nonlinear distortion degradation and insufficient long-term stability are solved at the same time.

The invention relates to the technical field of MEMS gyroscope signal processing, and discloses a deep learning dynamic noise reduction method for an MEMS gyroscope, and the method comprises the steps: obtaining original angular velocity data and background noise statistical data, carrying out the dynamic window adjustment and frequency domain separation processing to obtain a preliminary noise reduction signal, and generating a final noise reduction output through a physical constraint compensation network. Wherein the dynamic window adjustment uses an angular velocity increment prediction model to adjust the sequence length; the frequency domain separation distributes thresholds according to the motion intensity; the physical constraint compensation network is combined with temperature data to eliminate errors. According to the method, signal and noise characteristics and physical factors are fully considered, noise can be effectively suppressed, the accuracy and reliability of output data of the MEMS gyroscope are improved, and the method is suitable for multi-scene application.

The invention relates to the technical field of voice processing and enhancement, in particular to a noise isolation and target voice enhancement system based on specific voice pre-storage, and the system extracts pre-stored features from a pre-stored target voice database module through collecting environment audio signals in real time and carrying out frame segmentation and frequency domain transformation; adaptive modulation of amplitude and phase of each sub-band signal is realized by combining energy sensing sub-band modulation, then environmental noise energy is identified through dynamic noise isolation and multi-stage suppression is carried out, and finally continuous and clear target voice output is generated through enhanced fusion. The method can achieve the effective enhancement and environmental noise suppression of the target voice in a complex and changeable environment, improves the voice recognition precision and definition, supports the dynamic switching of multiple teachers and multiple classes, adapts to the target voiceprint change, and achieves the high-robustness voice enhancement and noise reduction in teaching, conference and multi-target scenes.

The invention discloses a weak fault sound source monitoring and early warning method based on deep learning and an acoustic array, and belongs to the technical field of equipment state monitoring and fault diagnosis. In order to solve the problem of early fault detection caused by easy submergence of weak fault voiceprint signals, incomplete feature extraction and scarcity of fault samples in a strong noise environment, the method comprises the following steps: collecting multichannel signals through an acoustic array, establishing a dynamic noise baseline by adopting a Gaussian mixture model, and performing noise suppression in combination with a spectral subtraction method; extracting a composite feature vector containing time domain, frequency domain and time-frequency domain features; and carrying out fault identification by using the pre-trained and fine-tuned transfer learning model, and finally outputting an early warning according to an identification result. The method is suitable for online monitoring and early warning of early weak faults of industrial equipment.

The invention relates to the field of Bluetooth headset control, in particular to an intelligent control method and device for a Bluetooth headset, equipment and a storage medium. The method comprises the following steps: acquiring multi-dimensional environment sensing parameters, performing scene noise analysis, and constructing scene sensing data; performing dynamic noise reduction parameter decision according to the scene perception data, and constructing a self-adaptive noise reduction strategy; collecting user physiological monitoring signals, performing real-time physiological state evaluation and user emotional state evolution, and constructing emotional state features; environment interference antagonism parameter optimization is carried out according to the emotional state characteristics, and spatial sound effect optimization data are constructed; according to the invention, the audio demand of the user is accurately met, corresponding adjustment is made in advance, and the user experience and the audio effect are improved.

The invention relates to the technical field of dam safety monitoring, and discloses a dam ant damage real-time monitoring and early warning method and system and a storage medium, and the method comprises the steps: constructing a dynamic noise feature space and a coupling feature model; sound vibration signals and multi-dimensional environment factor data are collected in real time; calling a noise space based on the environment data, and carrying out adaptive filtering on the sound vibration signal; analyzing the noise-reduced signal and environment data by using a coupling model to obtain a preliminary risk level; according to the grade, active acoustic exploration is selectively started for physical verification, or reverse optimization of the noise model is executed; and finally, the risk level is updated according to the verification result and early warning is output. Through combination of environment adaptive filtering and active verification, the problems of low monitoring accuracy and poor reliability caused by incapability of suppressing strong time-varying noise and lack of reliable verification means for suspected signals in the prior art are solved, and high-precision and high-reliability intelligent monitoring and early warning for dam ant damage are realized.

An SRAM architecture is disclosed that provides a process- and temperature-compensated word-line underdrive scheme enabling stable, low-voltage operation. Each row decoder generates an initial word-line signal, derives an inverse word-line signal, and drives the corresponding word line through driver circuitry including a word-line underdrive p-channel transistor. The transistor's gate is controlled by the inverse word-line signal and optionally receives a dynamically generated negative bias to reduce device size while maintaining performance. Control circuitry includes a word-line underdrive sink circuit having dummy memory cells that replicate bit-cell behavior to generate a compensated driver output signal tracking process and temperature variations. A negative bump generator produces a transient below-ground potential at the gate node to enhance underdrive efficiency. This improves bit-cell stability, dynamic noise-margin yield, and performance uniformity across process corners and temperature, while minimizing silicon area.

The application provides a kind of active noise reduction control method based on smoke stove linkage, range hood, electronic equipment and storage medium, belong to range hood technical field, wherein method includes obtaining the firepower grade signal of the stove;According to the corresponding relationship between the preset stove grade and the noise reduction parameter, determine the noise reduction parameter corresponding to the firepower grade signal;According to the noise reduction parameter, control the active noise reduction module to emit reverse sound wave.The application makes the reverse sound wave can match with the noise under different grades, realizes the accurate adaptation of noise reduction parameter and dynamic noise.

This invention discloses a dynamic noise active suppression suspension system for air-to-ground electromagnetic detection, belonging to the field of geophysical exploration technology. It includes a coil vibration suppression system connected to a UAV, a receiving coil sensor, and a coil attitude control system. The coil vibration suppression system is connected to the pitch frame of the coil attitude control system via a carbon fiber tube assembly. The pitch frame is connected to the receiving coil sensor via two roll shafts. One roll shaft is connected to a first motor, and a second motor is mounted on the frame edge of the pitch frame perpendicular to the roll shaft. Both the first and second motors are connected to a controller. The coil vibration suppression system suppresses the vibration amplitude of the receiving coil and absorbs the motion noise generated by the receiving coil vibration. The controller collects and controls the first and second motors to adjust the attitude of the receiving coil, suppressing the motion noise generated by the attitude change of the receiving coil. This application significantly reduces the motion noise of air-to-ground transient electromagnetic reception data, enabling low-noise detection using air-to-ground transient electromagnetic methods in complex environments.

The application discloses a call voice real-time noise reduction method and system based on dynamic noise perception, acquires a noisy original voice signal, carries out frame division and windowing pretreatment on the digital voice signal with noise, extracts the spectral characteristics of the noise, divides the noise into steady-state noise, non-steady-state noise and burst noise according to the spectral characteristics of the noise, constructs a lightweight neural network structure through model pruning and quantization, adjusts the parameters and strategies of the types of the steady-state noise, the non-steady-state noise and the burst noise to realize optimized reduction, carries out post-processing such as window removal and overlap addition on the voice signal after noise reduction, and outputs the voice signal after noise reduction; the application effectively removes the background noise in a complex noise environment through dynamic noise perception and an adaptive noise reduction strategy, significantly improves the voice quality, meets the real-time requirement, realizes real-time noise reduction processing under low delay conditions through lightweight model design, and meets the real-time requirement of voice communication and voice control.

Embodiments of the present application disclose a kind of group adaptive integrated control method, device, electronic equipment and storage medium, involve embodied intelligent technical field, wherein, the method includes: according to target scene characteristic, utilize physical modeling and domain randomization technology to generate diversified environment task set, and pre-processing is carried out to ensure data quality, train reinforcement learning coach based on these tasks, combine decoupling and backtracking policy distillation technology to synthesize training sequence, enhance the generalization and adaptive ability of model, adopt long-time memory model structure design base model, integrate multi-agent observation module to capture time sequence and spatial correlation, dynamic noise injection and asynchronous optimization are carried out in training process, the trained field base model is deployed to target scene, form collaborative network, and set fault-tolerant mechanism to deal with agent offline and other sudden situations, continuously collect actual data to optimize model parameters.The present application effectively solves the problems of high customization, weak migration and high maintenance cost in the prior art.

The application discloses a 12-lead electrocardiosignal denoising method and system based on a dual generative adversarial network, which comprises the following steps: obtaining time-adjacent clean electrocardiosignal segments and noisy electrocardiosignal segments; constructing a noise generative adversarial network, wherein the generator takes a random noise vector as input and simulates noise as output, the discriminator takes the noisy electrocardiosignal segment as a positive sample and a synthetic signal composed of the clean electrocardiosignal segment and the simulated noise as a negative sample for adversarial training, and a noise generator snapshot library is generated; performing dynamic noise injection on the clean electrocardiosignal segment to obtain 'noisy-clean' paired data; constructing a denoising network comprising a denoiser; obtaining the 12-lead electrocardiosignal containing noise and inputting it into the denoiser to output the denoised 12-lead electrocardiosignal. The application aims to overcome the defects in the prior art, such as the lack of real 12-lead electrocardiosignal denoising paired data and the poor model generalization ability caused by excessive dependence on the Gaussian noise assumption.

A voice capture system is provided that includes a first voice beamformer and a second voice beamformer, a voice mixer, a voice-rejecting noise beamformer, a noise beamformer adjuster, an interference suppressor, and a speech enhancer. The first voice beamformer and the second voice beamformer and the voice mixer generate a voice-enhanced reference signal based on a first frequency-domain microphone signal and a second frequency-domain microphone signal. The voice-rejecting noise beamformer includes filter weights and generates a noise reference signal based on the first frequency-domain microphone signal and the second frequency-domain microphone signal. The noise beamformer adjuster adjusts one or more filter weights of the voice-rejecting noise beamformer to account for fit variation. The interference suppressor generates an interference-suppressed signal based on the voice-enhanced reference signal and the noise reference signal. The speech enhancer dynamically generates an output voice signal by applying a dynamic noise suppression signal to each frequency bin of the interference-suppressed signal.