1663 results about "Signal characteristic" patented technology

The invention discloses a data driving-based power battery safety risk assessment method and system, particularly relates to the technical field of battery safety monitoring, and is used for solving the problems of false alarm and missing detection caused by confusion of electrochemical relaxation effect and real fault signal characteristics in the existing standing stage fault detection method. Through multi-source data acquisition based on voltage, temperature and state of charge in a standing stage, a current slope extreme point is identified through charge and discharge mode weighted correction, a time-delay mapping relation with a voltage curve geometrical morphology abrupt change point is established, abnormal inflection points are screened, and fractal dimension attenuation characteristics are quantified; according to the method, abnormal distribution density and material damage indexes are fused to generate a dynamic threshold value, composite criterion analysis is carried out by combining frequency domain energy offset characteristics extracted by time-frequency transformation and heat conduction time-lag phase difference, relaxation interference signals are effectively stripped, hidden danger characteristics are locked, full-dimension safety state evaluation from microdefect evolution to macroscopic parameter abnormity is realized, and the safety state evaluation accuracy is improved. And the reliability of power battery fault detection under complex working conditions is improved.

The invention discloses a power distribution network line fault positioning detection system, and relates to the technical field of power distribution network fault detection. The system comprises a mixed information acquisition layer, a fault feature extraction layer, an intelligent diagnosis layer and a fault positioning layer. The mixed signal acquisition layer comprises a high-frequency transient wave recording unit, a power frequency measurement unit, a wireless pulse sensor and a distributed optical fiber temperature measurement unit; the fault feature extraction layer comprises a time-frequency analysis module, a preprocessing module and a three-dimensional feature vector module; the intelligent diagnosis layer comprises a convolutional attention network, a space-time diagram neural network and a transfer learning module; the fault positioning layer comprises a particle swarm module and a fuzzy reasoning module. According to the invention, data information of the cable is acquired through the mixed information acquisition layer, a video analysis window function is dynamically matched with signal characteristics, a time domain graph scale, a frequency domain resonance component and a space field intensity gradient are constructed, fault diagnosis and positioning are carried out by using the intelligent diagnosis layer, and the fault positioning detection efficiency of the power distribution network is improved.

The invention relates to the technical field of low-altitude safety emergency, and discloses a low-altitude safety emergency system based on large model application. The system comprises a low-altitude situation awareness module which constructs a dynamic situation prediction model based on historical flight data, collects a low-altitude target track coordinate sequence, radar echoes and communication signal characteristics in real time, and outputs a theoretical security situation value. The multi-source data fusion analysis module performs three-dimensional difference analysis on theoretical and actually measured situation values, which relates to time domain trend deviation, space domain coverage offset and event sequence matching degree, and generates a target-level difference feature matrix; and the risk propagation path deduction module inputs the matrix into a geographic topology analysis network, generates a probability thermal distribution diagram of an abnormal risk propagation path in combination with airspace node communication parameters and a target position, and locates a risk influence area. And the intelligent disposal strategy generation module configures parameters according to the graph, starts high-frequency monitoring for a high-probability area, and performs communication disturbance test on an adjacent target.

The invention relates to the technical field of computer-aided communication, in particular to a wireless mouse strong interference source avoiding method based on a wireless frequency hopping model, which comprises the following steps: acquiring multi-band signal characteristics by monitoring electromagnetic spectrum data of an environment where a wireless mouse is located in real time; constructing an interference source analysis model by using an improved weighted fuzzy clustering algorithm, and accurately identifying the type and intensity distribution of an interference source; according to an analysis result, dynamically generating a frequency hopping strategy by using a channel state prediction model based on reinforcement learning, and determining priority sequences of a main channel and a standby channel; when the interference on the main channel exceeds a preset threshold value, performing real-time evaluation by means of a composite quality index and switching to an optimal standby channel; meanwhile, through a closed-loop feedback mechanism, according to the difference between actual channel quality and expected channel quality, a stochastic gradient descent method is adopted to optimize frequency hopping strategy parameters, and dynamic self-adaptive adjustment of interference avoidance is achieved. Complex electromagnetic environment interference can be effectively dealt with, and the communication stability of the wireless mouse and the use experience of a user are remarkably improved.

The invention discloses an online multi-channel rapid gas chromatographic analysis system and method, particularly relates to the technical field of gas chromatographic analysis, and is used for solving the problems of baseline noise rise and trace component detection precision reduction caused by physical coupling and signal interference when multiple detectors work in parallel in the prior art. Interference signal features of adjacent detection channels are extracted, a chaotic or linear compensation signal is dynamically generated in combination with harmonic primary and secondary sorting and recursive quantization factor analysis, and common-mode interference of a target channel is suppressed. Electromagnetic isolation parameters of a shielding structure are adjusted in real time based on the baseline signal after noise suppression, and accurate matching of impedance and interference spectrum is achieved; further separating time domain overlapped signals and analyzing asymmetric distortion characteristics, and eliminating system errors through normalization calibration; dynamic interference suppression, electromagnetic shielding optimization and signal distortion traceability are combined, the anti-interference capability and analysis precision of multi-channel detection are remarkably improved, and the method is suitable for high-throughput rapid detection scenes of complex samples.

The invention belongs to the technical field of signal acquisition and processing, and discloses a low-noise biopotential signal acquisition system and a low-noise biopotential signal processing method. An original bioelectricity signal of a human body is collected through an electrode array, and analog-to-digital conversion is carried out after the signal is processed by a multi-stage self-adaptive filtering and amplifying circuit. A digital signal is sequentially subjected to adaptive wavelet transform denoising and empirical mode decomposition to obtain a multi-level intrinsic mode function set, an intrinsic mode function of a characteristic frequency band is extracted from the multi-level intrinsic mode function set, and the signal is reconstructed. Introducing a dual noise reduction mechanism combining adaptive wavelet transform and empirical mode decomposition; a signal quality real-time evaluation system is established, and system parameters including the gain of a variable gain amplifier and the bandwidth of a dynamic band-pass filter are dynamically adjusted through closed-loop feedback. The self-adaptive optimization of the whole biopotential signal acquisition process is realized, various interferences are effectively inhibited, the signal characteristics are reserved, and the signal-to-noise ratio is remarkably improved.

The invention relates to a GIS partial discharge digital twinborn identification method based on a simulation fault database and a storage medium. The problems that in the prior art, due to the fact that GIS partial discharge signal features have large randomness, the accuracy and precision of algorithm recognition and positioning are not high, meanwhile, manpower and material resources are high in cost, complex operation conditions are difficult to simulate, and the defect type coverage range is not wide are solved, and the adopted technical scheme is that a GIS three-dimensional geometric model is constructed based on a GIS true model; setting a digital discharge source in the GIS three-dimensional geometric model, discharging through the digital discharge source, obtaining UHF signals in batches, and extracting time domain pulse waveform characteristic parameters in the UHF signals; constructing a simulation fault database based on the time domain pulse waveform characteristic parameters; constructing a fault inversion network based on the simulation fault database; recognizing and positioning a fault through a fault inversion network; based on the GIS partial discharge digital twinborn model, a fault identification result and a positioning process are arranged on the GIS partial discharge digital twinborn model, and fault diagnosis and fault location positioning can be accurately and visually carried out.

The invention discloses a head-mounted brain body workability assessment method based on a space-time dynamic fusion network, belongs to the technical field of artificial intelligence and human body signal acquisition, and solves the problems that an existing method cannot comprehensively reflect the comprehensive state of an individual, and interaction and potential risks among multiple factors are difficult to reveal. The method comprises the steps that physiological parameter signals are collected based on wearable equipment, and an attention and fatigue prediction result is output by an LSTM prediction model; according to the method, the graph structure data packet is analyzed and processed based on the pre-constructed space-time dynamic fusion network model, the working state signal characteristics of the plateau high-risk operating personnel can be accurately evaluated, the space-time graph neural network (STGNN) can capture the complex relation, the high-dimensional topological characteristics are extracted, and the high-risk working personnel can be accurately evaluated. The dynamic event-driven pulse neural network (deSNN) can convert the pulse sequence features into pulse sequence features so as to describe feature values more accurately, iterative optimization enables the pulse sequence features and the pulse sequence features to cooperate, and the accuracy and reliability of time-space dynamic fusion network model evaluation are improved.

Accordingly, there are disclosed herein receivers and receiving methods that provide a graceful transition from PAM2 to PAM4 signaling. One illustrative method includes: negotiating a link speed having PAM4 signaling; performing adaption of at least one gain or filter coefficient during PAM2 signaling; switching to PAM4 detection before receiving PAM4 signaling; disabling said adaptation before said switching to PAM4 detection; detecting PAM4 signaling using at least one statistic of detected PAM4 symbols; and enabling said adaptation after PAM4 signaling is detected. Another illustrative method includes: negotiating a link speed having PAM4 signaling; adapting at least one of gain and filter coefficients during PAM2 signaling; monitoring for a change in at least one signal characteristic while performing PAM2 detection; and transitioning to PAM4 detection after detecting said change.

The invention relates to the technical field of ultrasonic flaw detection of titanium alloy bars, in particular to an ultrasonic flaw detection method and system for titanium alloy bars, and the method comprises the following steps: obtaining ultrasonic signals of all positions; obtaining a receiving duration sequence of each position; according to the difference between adjacent peak values in the ultrasonic signal of each position, the dispersion degree of all the peak values, the average level of the receiving duration sequence and the difference between the receiving duration sequence of each position and the receiving duration sequence of the adjacent position, the real defect of each position is obtained; according to the dispersion degree of the amplitude average level of all the modal components of each position, the total number of the modal components and the energy concentration degree and the bandwidth of the ultrasonic signal of each position in the frequency domain, obtaining a signal abnormal factor of each position; defect factors of all the positions are obtained, and then whether defects exist in all the positions or not is judged. By analyzing the signal characteristics of the real defects in the titanium alloy bar, the accuracy of ultrasonic flaw detection of the titanium alloy bar is improved.

The invention relates to the technical field of circuit boards, and discloses a circuit board interlayer connection detection method, device and equipment, and the method comprises the steps: carrying out the electrical characteristic parameter measurement of a multilayer flexible circuit board, and obtaining the characteristic parameter data of the circuit board; designing and arranging a multi-test-point detection structure on the multi-layer flexible circuit board, and performing signal excitation to obtain test point position information; applying a differential excitation signal, and collecting a differential response waveform between the test points to obtain a time domain reflection signal feature; impedance spectrum conversion and parameter extraction are carried out on the time domain reflection signal features, and an interlayer connection impedance distribution diagram is obtained; and establishing an interlayer connection defect characteristic model aiming at nickel-based alloy wire fracture, interface cracking and liquid metal fiber felt delamination in various bending states by utilizing the interlayer connection impedance distribution diagram, so that accurate classification of defect types and severity is realized, and interlayer connection faults and surface welding quality problems can be distinguished; and the misjudgment rate is greatly reduced.

The invention provides a transformer on-load tap-changer state evaluation method and system, electronic equipment and a medium, and belongs to the field of fault diagnosis of transformer on-load tap-changers. The method comprises the following steps: acquiring electrical and mechanical performance parameters of the on-load tap-changer of the transformer, and generating an oil chromatographic data characteristic matrix according to oil chromatographic data; for the vibration signals, generating a vibration signal feature matrix; generating a driving motor current signal characteristic matrix according to the driving motor current signal; fusing the three feature matrixes to generate a multi-dimensional information feature matrix; and calculating correlation among data by using an attention mechanism, endowing a multi-dimensional information feature matrix with a weight, and inputting the multi-dimensional information feature matrix into a BIGRU model for state evaluation. The characteristic data of the electrical characteristics and the mechanical characteristics of the on-load tap-changer are comprehensively selected, and the evaluation model is used for fault diagnosis, so that the diagnosis accuracy and adaptability are improved.

The invention provides a hydro meteorology early warning method for an offshore oil and gas platform, and belongs to the technical field of offshore hydro meteorology. Extreme weather events are identified by adopting minimum probability abnormal event identification vectors to match abnormal characteristic parameters, and abnormal signal characteristic parameters are input into an ocean dynamics prediction model to calculate real-time sea condition parameters; calling a multi-temporal-spatial-scale early warning fusion matrix to combine with a wavelet decomposition technology and a recurrent neural network to realize multi-scale information integration, analyzing an environmental parameter change trend through a sea condition jump identification model and triggering an emergency response, dynamically adjusting system parameters according to a stability evaluation index vector, and optimizing prediction precision by adopting an early warning residual value compensation matrix. And finally, multi-level early warning information is generated and a real-time early warning notification is sent to an operator, so that the technical problem of insufficient early warning precision of an offshore oil and gas platform hydro meteorology early warning system in multi-spatio-temporal scale data fusion processing is solved.

A distributed optical fiber sensing event classification and identification method based on mask auto-encoder self-supervised learning belongs to the technical field of distributed optical fiber sound and vibration sensing, on one hand, a pre-training network does not need label data, a downstream classification and identification network only needs a small amount of labeled data, and the demand of the labeled data is reduced; and on the other hand, in the pre-training process, the model extracts DAS signal features and compresses the calculation amount of the DAS signals, so that pixel-level reconstruction of DAS signal reconstruction is changed into feature reconstruction, a large amount of time is saved, and the accuracy is ensured.

The invention provides a rotary mechanical equipment fault identification method and device based on vibration data, and belongs to the field of vibration monitoring. The method comprises the following steps: acquiring a component information library of to-be-monitored rotary mechanical equipment; fault information corresponding to the component information base is determined, the fault characteristic frequency of each fault is calculated, and a fault characteristic frequency model corresponding to the component information base is established; collecting a vibration signal characteristic value of each component in the component information base in real time; extracting a vibration signal characteristic value, and matching the frequency domain characteristic value with the fault characteristic frequency based on a component information base and a fault characteristic frequency model to obtain a matching result; positioning an abnormal feature in the signal based on the time domain feature value and the geometric feature value; and identifying the fault type and degree of the mechanical equipment according to the matching result and the abnormal features. According to the mechanical equipment vibration monitoring data processing method and device, potential faults of the equipment can be found as soon as possible, and the equipment can be maintained in time.

The invention provides an electric energy meter PCB function test method and device, a medium and equipment, and the method comprises the steps: obtaining a multi-channel signal source, automatically adjusting the characteristics of an input signal based on a feedback signal of an electric energy meter through a self-adaptive algorithm, and determining the consistency of the working characteristics of the input signal and the working characteristics of the electric energy meter; a virtual power grid environment simulation model is constructed, through integrated power simulation software and dynamic interference simulation, simulation signal source automatic adjustment signal characteristics are input into the virtual power grid environment simulation model based on the integrated power simulation software to synthesize a composite signal, and load fluctuation of an electric energy meter in actual operation is simulated based on the composite signal. Determining an electric energy meter test data set; according to the invention, the fault mode of the electric energy meter can be automatically identified, the test data can be analyzed in real time, and fault diagnosis and optimization suggestions can be automatically given. The intelligent diagnosis mode is far better than a traditional manual detection method, and the speed and accuracy of fault diagnosis can be remarkably improved.

The invention relates to the technical field of submarine engineering and information processing, in particular to a submarine cable positioning method and system, and the method comprises the steps: synchronously collecting magnetic field, structured light and underwater acoustic signal data, and fusing inertial navigation information to construct a unified space-time reference system; precise registration of heterogeneous data is realized by adopting sliding time window screening, multi-modal noise reduction and rigid body transformation; constructing a graph structure model for expressing cable space-signal features based on the fused data, and reasoning cable path topology by using a graph convolutional network; reconstructing a cable center line through three-dimensional visualization, embedding the cable center line into submarine topography, and realizing risk area dynamic clustering by combining signal integrity score and burial depth estimation; historical time sequence data are fused, a long short-term memory network is adopted to predict cable drift, detectable attenuation and burial depth change trends, risk levels are divided based on prediction results, and spatial distribution early warning and disposal suggestions are output. According to the invention, the cable positioning precision and the risk assessment intelligent level in a complex seabed environment are improved.

A fault identification system and method. The system comprises an edge device and a server. The edge device can extract, from an original vibration signal, signal characteristics reflecting an abnormal vibration state of a device under test and report the signal characteristics to the server; and the server can identify a fault state of the device under test on the basis of the signal characteristics. The data volume of the signal characteristics reported in the process is small, thereby reducing the bandwidth requirement; and the signal characteristics can be used for accurate fault identification, thereby improving the accuracy of fault identification.

The invention relates to the technical field of power equipment inspection, in particular to a high-voltage electrical equipment inspection method and system based on microphone array measurement. The method comprises the following steps: acquiring array sound signal data, analyzing the array sound signal data, and determining sound signal feature information; obtaining a layered structure parameter set and an insulating medium parameter set of the target electrical equipment, performing targeted compensation and correction on the sound signal feature information according to the layered structure parameter set and the insulating medium parameter set, and determining abnormal voiceprint feature information; and generating abnormal component information according to the abnormal voiceprint feature information, and determining and outputting an abnormal detection report. According to the invention, non-contact real-time detection of the high-voltage electrical equipment in operation is realized, the adaptability of the detection process to the high-voltage electrical equipment which is internally provided with a multi-layer composite structure and is filled with an insulating medium is enhanced, and the positioning accuracy of a fault position in the equipment is improved.

The invention discloses an audio signal processing method and device based on MOE, a medium and a product, and relates to the field of audio signal processing. According to the technical scheme, firstly, the audio signal processing device decomposes the audio signal into the sub-band signals in different frequency intervals, and the complexity parameter of each sub-band signal is calculated; on the basis of the sub-band signals and complexity parameters corresponding to the sub-band signals, the gating network in the hybrid expert model can intelligently distribute the optimal expert network combination for each sub-band signal, the speciality of each expert network is fully utilized, the processing efficiency is improved, the optimal expert network combination is screened by adopting a processing weight coefficient threshold, and the processing efficiency is improved. And redundant processing is avoided. And finally, the audio signal processing device obtains a structured audio data stream through quantization coding and data packaging. According to the dividing and treating method, the quality and efficiency of audio signal processing are improved, the processing strategy can be flexibly adjusted according to the signal characteristics, and the adaptability is high.

The invention provides a wireless signal detection and positioning method and system, and relates to the technical field of wireless communication, and the method comprises the steps: collecting signal characteristic parameters through a plurality of monitoring devices, constructing a three-dimensional time-frequency space matrix, carrying out envelope modulation to construct a primary function, carrying out iterative projection to obtain a reconstructed signal, and extracting a characteristic fingerprint through a high-order statistical moment; constructing a topological graph according to geographic distribution to calculate information entropy, and performing dynamic clustering fusion; a scene type is identified and a location is estimated in combination with historical data. The wireless signal detection accuracy and the positioning precision can be improved, and the method adapts to a complex electromagnetic environment.

The disclosure generally relates to a method, system and apparatus to improve a user's understanding of speech in real-time conversations by processing the audio through a neural network contained in a hearing device. The hearing device may be a headphone or hearing aid. In one embodiment, the disclosure relates to an apparatus to enhance incoming audio signal. The apparatus includes a controller to receive an incoming signal and provide a controller output signal; a neural network engine (NNE) circuitry in communication with the controller, the NNE circuitry activatable by the controller, the NNE circuitry configured to generate an NNE output signal from the controller output signal; and a digital signal processing (DSP) circuitry to receive one or more of controller output signal or the NNE circuitry output signal to thereby generate a processed signal; wherein the controller determines a processing path of the controller output signal through one of the DSP or the NNE circuitries as a function of one or more of predefined parameters, incoming signal characteristics and NNE circuitry feedback.

The disclosure generally relates to a method, system and apparatus to improve a user's understanding of speech in real-time conversations by processing the audio through a neural network contained in a hearing device. The hearing device may be a headphone or hearing aid. In one embodiment, the disclosure relates to an apparatus to enhance incoming audio signal. The apparatus includes a controller to receive an incoming signal and provide a controller output signal; a neural network engine (NNE) circuitry in communication with the controller, the NNE circuitry activatable by the controller, the NNE circuitry configured to generate an NNE output signal from the controller output signal; and a digital signal processing (DSP) circuitry to receive one or more of controller output signal or the NNE circuitry output signal to thereby generate a processed signal; wherein the controller determines a processing path of the controller output signal through one of the DSP or the NNE circuitries as a function of one or more of predefined parameters, incoming signal characteristics and NNE circuitry feedback.

The invention provides a full-focusing ultrasonic imaging method for multi-modal damage of a composite material, and the method comprises the steps: calculating the correlation of the reflection intensity of a sound wave signal at different moments and spatial positions through employing an established quantitative corresponding relation, recognizing a close contact region and a loose separation region between a bonding layer and a substrate, and carrying out the recognition of the close contact region and the loose separation region. Judging the continuity distribution condition of the interface contact state; according to the continuous distribution condition of the interface contact state, key feature points of dynamic damage evolution are extracted, a time sequence analysis method is adopted to track the feature points to adapt to damage features of dynamic changes, and prediction data of the damage expansion trend are obtained; and according to the dynamic representation result of the interface state, fusing the obtained multi-period signal characteristic data, carrying out quantitative calculation on the area change of the debonding region, judging the residual distribution state of the bonding strength, and obtaining a comprehensive evaluation report of the damage degree of the bonding interface.

The invention discloses a multi-probe on-line real-time film thickness measuring method and a multi-probe on-line real-time film thickness measuring device. In combination with the multi-wavelength analysis capability and multi-channel parallel signal characteristics of the spectrum detection unit, different plane positions are adopted to correspond to different wavelengths or central wavelengths, the transmissivity of each wavelength is obtained, and the thickness of the corresponding position is obtained respectively; the precision and the accuracy of obtaining the thickness by a single wavelength are improved through a multi-mode cooperative signal analysis algorithm, so that the thicknesses of a plurality of different positions of the same thin film are accurately obtained at the same time through one spectrum detection unit, or the thicknesses of a plurality of different thin films are obtained; according to the invention, the problems of large absorption, transparent or non-transparent substrate compatibility, quick response, multi-branch and on-line real-time measurement of the film are successfully solved; the device and the method can be integrated in film growth equipment to carry out rapid and high-precision measurement on the film, so that closed-loop control on film production is realized, the iteration speed is increased, the yield of high-quality film production is greatly increased, and the film production cost is saved.

The disclosure generally relates to a method, system and apparatus for processing audio through a neural network contained in a hearing device. In one embodiment, the disclosure relates to an apparatus to enhance incoming audio signal. The apparatus includes a controller to receive an incoming signal and provide a controller output signal; neural network engine (NNE) circuitry in communication with the controller, the NNE circuitry activatable by the controller, the NNE circuitry configured to generate an NNE output signal from the controller output signal; and digital signal processing (DSP) circuitry to receive one or more of controller output signal or the NNE circuitry output signal to thereby generate a processed signal; wherein the controller determines a processing path of the controller output signal through one of the DSP or the NNE circuitries as a function of one or more of predefined parameters, incoming signal characteristics and NNE circuitry feedback.

A connected device is provided. The connected device comprises processing circuitry configured to determine that a wireless transmission originated from a medical device by inspecting the wireless transmission to identify a signal characteristic, header data, or payload data of the wireless transmission, and determining that the signal characteristic, header data, and / or payload data of the wireless transmission matches a predetermined medical device signal characteristic, header data, and / or payload data. The processing circuitry is further configured to, upon determining that the wireless transmission originated from a medical device, perform medical device-specific processing on the wireless transmission.

The invention relates to the technical field of cutting machine control, in particular to a wire cutting machine feeding speed control system and method.The wire cutting machine feeding speed control method comprises the steps that a multi-mode signal recognition material module extracts current and vibration signal features through the self-adaptive feature selection technology, and the current and vibration signal features are obtained on the basis of an improved random forest model; a processing material is identified by combining dynamic weight voting and a weighted fusion method, and online learning and model updating are supported so as to adapt to processing requirements of different materials. The double-loop self-adaptive control module comprises outer-loop initial compensation and inner-loop fuzzy PID adjustment, real-time fine adjustment parameters are optimized according to historical machining records and double-target error values, machining precision and stability are ensured, and machining quality is improved. And the energy consumption management optimization module intelligently regulates and controls the energy consumption of the system by identifying non-processing and processing periods, so that the overall energy efficiency is improved, and energy-saving production is realized. According to the invention, the efficiency and the accuracy of the wire cutting machine in machining of multi-variety and small-batch dies are improved.

The disclosure generally relates to a method, system and apparatus to improve a user's understanding of speech in real-time conversations by processing the audio through a neural network contained in a hearing device. The hearing device may be a headphone or hearing aid. In one embodiment, the disclosure relates to an apparatus to enhance incoming audio signal. The apparatus includes a controller to receive an incoming signal and provide a controller output signal; a neural network engine (NNE) circuitry in communication with the controller, the NNE circuitry activatable by the controller, the NNE circuitry configured to generate an NNE output signal from the controller output signal; and a digital signal processing (DSP) circuitry to receive one or more of controller output signal or the NNE circuitry output signal to thereby generate a processed signal; wherein the controller determines a processing path of the controller output signal through one of the DSP or the NNE circuitries as a function of one or more of predefined parameters, incoming signal characteristics and NNE circuitry feedback.

The disclosure generally relates to a method, system and apparatus to improve a user's understanding of speech in real-time conversations by processing the audio through a neural network contained in a hearing device. The hearing device may be a headphone or hearing aid. In one embodiment, the disclosure relates to an apparatus to enhance incoming audio signal. The apparatus includes a controller to receive an incoming signal and provide a controller output signal; a neural network engine (NNE) circuitry in communication with the controller, the NNE circuitry activatable by the controller, the NNE circuitry configured to generate an NNE output signal from the controller output signal; and a digital signal processing (DSP) circuitry to receive one or more of controller output signal or the NNE circuitry output signal to thereby generate a processed signal; wherein the controller determines a processing path of the controller output signal through one of the DSP or the NNE circuitries as a function of one or more of predefined parameters, incoming signal characteristics and NNE circuitry feedback.