26 results about "Speech clarity" patented technology

The utility model relates to the technical field of wall surface cloth, in particular to curved surface acoustic wall surface sticking cloth. The patch comprises a patch body, two convex grooves are formed in the surface of the patch body, strip-shaped plates are connected to the surfaces of the convex grooves in a sliding mode, connecting blocks are fixedly connected to the surfaces of the strip-shaped plates, flannelette is fixedly connected to the surfaces of the two connecting blocks, and the flannelette is attached to the surface of the patch body. The surface of the flannelette is fixedly connected with a plurality of flannelette points, the height sizes of the flannelette points are different, the surface of the patch body is provided with a plurality of sound absorption grooves, and a plurality of refraction protrusions are arranged in the sound absorption grooves. The problems that when sound waves meet a flat patch body, the sound waves on the surface of the patch body are intensively reflected, so that indoor sound interferes with each other, echo is obvious, voice definition is reduced, noise pollution is aggravated, normal communication, learning and rest of people are seriously affected, and even hearing fatigue is possibly caused are solved.

The present invention provides an audio processing integration method and device for a hearing aid. The method includes: acquiring in real time an audio signal collected by a hearing aid microphone; processing the audio signal through a feature extraction network to obtain first audio feature data; performing frequency domain conversion, noise distribution prediction and weighted filtering processing on the first audio feature data based on a noise suppression network to obtain first noise-reduced audio features; processing the first audio feature data by using an adaptive enhancement network, and obtaining second enhanced audio features by randomly selecting frequency reference points, collecting feature points and processing; fusing the first noise-reduced audio features and the second enhanced audio features through a cross-attention network to obtain fused audio features; and finally processing the fused audio features by using an audio enhancement network to obtain an enhanced audio signal and outputting it. The present invention can improve the speech clarity and sound quality performance of a hearing aid in a complex environment and improve the auditory experience of hearing-impaired users.

The invention relates to the technical field of adaptive signal processing, in particular to an adaptive gain and noise reduction hearing aid based on artificial intelligence. The method comprises the following steps: receiving an acoustic signal through a double-microphone array, separating target voice from environmental noise through time delay inequality, and establishing a space sound field model; converting the audio signal into a time-frequency domain, and extracting an MFCC feature; a CNN-LSTM hybrid network model is constructed, and voice and noise are classified in real time; dynamically adjusting gain values of different frequency bands according to the hearing loss curve and the real-time signal-to-noise ratio; noise in a non-target direction is suppressed, and the spatial selectivity of the voice signal is enhanced; and optimizing model parameters according to the hearing archive of the user, and supporting individual difference adaptation. Acoustic signals are captured through a dual-microphone array, and target voice and environmental noise are distinguished by using delay inequality. A target voice signal is effectively separated, noise from a non-target direction is suppressed, and voice definition is improved.

The invention relates to a wind noise quality evaluation method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining the wind noise data and subjective evaluation results of a plurality of sound collection devices under each test condition; based on a preset evaluation standard, performing sound quality data analysis on the wind noise data of each test working condition to obtain a wind noise quality data analysis result of each test working condition, thereby determining a problem frequency band of each test working condition; and finally, obtaining a wind noise quality evaluation result of each test working condition according to the problem frequency band of each test working condition and the subjective evaluation result of each test working condition. Therefore, the problem that in the prior art, wind noise is directly evaluated by mostly using indexes such as sound pressure level and language definition, and technical means for evaluation from the angle of sound frequency characteristics are lacked is solved, the wind noise level of the vehicle under different working conditions can be systematically evaluated, and the evaluation accuracy is improved. And sound quality evaluation indexes corresponding to subjective evaluation can be found for noise with different frequencies.

The invention discloses an AI digital human device and method based on voiceprint discrimination and capable of communicating with multiple people face to face, sound signals from different directions are received through a microphone array, the distance between a speaker and the device is accurately measured through an infrared distance measuring sensor, and meanwhile, face tracking is carried out by means of a micro camera, so that the distance between the speaker and the device is accurately measured. And the position and the eye position of each speaker can be accurately captured. And secondly, through the combination of a beam forming algorithm and an RNNoise noise reduction module, the device can accurately extract the sound of a target speaker from a complex environment, remove background noise and improve voice definition. Through efficient extraction of the Mel frequency cepstrum coefficient and the ECAPA-TDNN deep learning model on the voice features, the system can realize high-precision voiceprint recognition, and through combination with a hash table storage mode, the voiceprint features of the speaker are quickly compared, so that the recognition accuracy is further enhanced.

The invention relates to the technical field of bone conduction vibrators, in particular to a bone conduction vibrator assembly special for hearing aid for improving tone quality, which comprises a vibration assembly and a vibration piece, and the vibration piece is of a concentric ring structure and sequentially comprises an inner ring, a middle ring and an outer ring from inside to outside. According to the bone conduction earphone, the frequency response range is widened and the voice definition is optimized through the gradient annular vibrating reed, resonance distortion is inhibited by combining with the elastic arm mode separation design, the disassembly and assembly efficiency is improved by matching with the split type modular shell, precise assembly is realized by assisting with the positioning convex seat, and the tone quality performance, the maintenance convenience and the structural reliability of the bone conduction earphone are comprehensively improved.

This specification provides one or more embodiments of an in-vehicle audio equalization method, apparatus, storage medium, and program product. The method includes: acquiring a background noise signal of the current in-vehicle environment and determining the gain difference spectrum of the background noise signal relative to a reference noise signal of a quiet environment; performing track-by-track processing on the currently playing media audio signal to obtain multiple independent audio tracks corresponding to different frequency ranges; for each independent audio track, determining a corresponding gain adjustment amount from the gain difference spectrum based on its frequency range, and applying a corresponding gain adjustment to the independent audio track when the gain adjustment amount meets a gain condition; merging the signals of each independent audio track to output as a processed media audio signal, thereby improving speech clarity and music listening experience while significantly improving the accuracy and naturalness of audio equalization.

The present invention discloses a method and system for identifying and calibrating spatial information of a hearing aid based on dual microphones. First, a dual microphone array pickup module is used to collect ambient sound signals and obtain a main microphone signal and a secondary microphone signal. Then, the main microphone signal and the secondary microphone signal are processed to obtain the maximum energy value of each frequency point of the two microphone signals. Next, the modulation amount of the maximum energy value of each frequency point at each frequency point is calculated. Based on the modulation amount of the maximum energy value of each frequency point at each frequency point; the present invention realizes the function of fully utilizing the spatial orientation information of the dual microphones of the hearing aid to identify different usage scenarios, and can refine and calibrate the algorithm architecture of the hearing aid according to the specific scenario so as to provide a better usage experience, significantly improve the speech clarity and wearing comfort, enhance the communication confidence and quality of life of users in daily life and social occasions, and is suitable for being widely promoted and used.

The application provides a three-dimensional virtual surround sound generation method and generation system, and relates to the field of surround sound. The application adopts an excellent structure of a howling elimination technology to eliminate howling, enhance spatial positioning and speech clarity. Reflection sound is constructed by infinite N-level echo simulation to construct a surround sound channel, so that the generated surround sound has more distant background sound, thereby making the virtual surround sound have strong spatial sense, distance sense, diffusivity and surrounding sense, rather than just positioning sense. A post sound channel is constructed by inversely superimposing the delayed signal on the direct signal, so that the spatial sense of the surround sound is obviously enhanced, and the sound seems to come from all directions. The original sound bass is preserved by separation technology, and the virtual sound in each direction is freely controlled, thereby controlling the sound intensity, tone brightness and tone softness.

The present invention discloses a real-time online detection and real-time feedback system for speech clarity, comprising a microphone for real-time acquisition of speech signals; an embedded computer for real-time short-time Fourier transform of an active speech signal extracted from the speech signal, dividing the obtained active speech spectrum into key frequency bands and calculating the spectrum envelope and average peak spectrum of each key frequency band, performing low-pass filtering on the average peak spectrum and then performing multi-resolution discrete Fourier transform, dividing the transformed result into fluctuation frequency bands and calculating the fluctuation spectrum mean of each fluctuation frequency band; calculating the speech modulation degree of the fluctuation frequency band according to the fluctuation spectrum mean, performing auditory correction on the speech modulation degree and then calculating the signal-to-noise ratio; calculating the transfer index, the modulation transfer index and the speech transfer index in sequence according to the signal-to-noise ratios of all fluctuation frequency bands; and displaying the speech transfer index as a speech clarity analysis result in a graphical manner.

The present disclosure provides a method and apparatus for training a multi-modal speech language large scale model, a device and a medium.SOLUTION: A solution is to input first question speech data into a multi-modal large speech language model to obtain first answer speech data generated by the multi-modal large speech language model, determine a question text corresponding to the first question speech data and an answer text corresponding to the first answer speech data, and determine a first score according to the question text and the answer text. A second score is determined according to a speech feature of the first question speech data and a speech feature of the first answer speech data, where the speech feature includes at least one of a speech clarity, a speech rate feature, a timbre feature, an intonation feature, and an emotion feature.SELECTED DRAWING: Figure 2

This utility model belongs to the field of hearing aid technology and discloses a behind-the-ear hearing aid, including a hearing aid suspension shell and an amplifier, receiver, and battery disposed inside the hearing aid suspension shell. Multiple miniature directional microphones are embedded on the front, top, and rear sides of the hearing aid suspension shell, and multiple independent acoustic channels are provided inside the hearing aid suspension shell. This utility model enhances the ability to pick up target speech from the front by embedding integrated miniature directional microphones on the front, top, and rear sides of the hearing aid shell in an asymmetrical spatial arrangement, and by providing multiple independent acoustic channels inside the hearing aid shell. Each independent acoustic channel is equipped with an electrically adjustable miniature acoustic damper, which can physically reduce noise, suppress side and rear noise, wind noise, and interference in specific frequency bands, and improve the basic signal-to-noise ratio. The independent acoustic channels can physically isolate the sound signals of each channel, improving speech clarity.

This invention relates to the field of electronic microphone audio signal processing technology, and provides an electronic microphone audio signal processing system and method, including a server connected to a signal acquisition module, a digital signal processing module, a digital-to-analog converter module, and an output module. The signal acquisition module includes a sound sensor, an analog front-end, and an analog-to-digital converter. The sound sensor uses a MEMS microphone array to convert changes in sound wave pressure into analog electrical signals. The analog front-end internally includes a preamplifier, a bias circuit, and an anti-aliasing filter. The preamplifier amplifies the electrical signal output by the microphone. The anti-aliasing filter filters out high-frequency components above the Nyquist frequency before analog-to-digital conversion, preventing aliasing distortion. This invention can process digital audio signals in real time and accurately, effectively eliminating echo, suppressing noise, stabilizing volume, and improving speech clarity and sound quality, greatly improving the audio signal processing quality of electronic microphones.

The invention belongs to the technical field of medical education, and discloses a method and a system for evaluating comprehensive clinical competency of medical students, video and audio data in an interaction process of the medical students and virtual patients are collected, a response text is obtained in combination with a voice recognition technology, and the comprehensive clinical competency of the medical students is evaluated through a retrieval enhancement generation technology. Performing semantic alignment on the response text of the medical student and the authoritative medical knowledge base; the method comprises the steps of evaluating language definition, common situation performance, attitude performance, eye contact rate and expression activeness of medical students on the basis of multi-modal data, finally, performing weighted fusion on various quantitative indexes to obtain a comprehensive score, generating a feedback report, and performing encrypted storage on original data; therefore, the problems of mutual separation of professional knowledge and clinical communication ability assessment, strong assessment subjectivity, lack of objective quantitative indexes and insufficient non-language communication assessment in medical student assessment in a current medical education system are solved.

The invention relates to an audio processing method, device and system, electronic equipment and a storage medium, and the method mainly comprises the steps: obtaining a first audio signal which is an audio signal obtained after an original audio signal is subjected to primary noise reduction and amplification; performing first residual noise detection on the first audio signal to obtain first detection result information associated with the first residual noise detection; under the condition that the first detection result information meets a preset secondary noise reduction and amplification starting condition, performing secondary noise reduction and amplification processing to obtain a second audio signal; and outputting the second audio signal. According to the technical scheme, whether secondary noise reduction amplification is started or not is determined through residual noise detection of the audio signal after primary noise reduction amplification, adaptive noise reduction amplification of the audio signal collected in a complex environment is achieved, and the voice definition degree of the audio signal collected in various environments can be improved.

The present invention discloses a method for predicting speech clarity, a computer device, and a storage medium. The method includes: performing a noise power test on a tire to obtain a noise power spectrum and a scanned road spectrum corresponding to at least one test condition; performing a first-band road noise simulation process on the noise power spectrum corresponding to each test condition to determine first road noise data corresponding to each test condition; performing a second-band road noise simulation process on the scanned road spectrum corresponding to each test condition to determine second road noise data corresponding to each test condition; performing an in-vehicle speech clarity analysis on the first road noise data and the second road noise data corresponding to each test condition to determine a target speech clarity corresponding to each test condition. This method can predict in-vehicle speech clarity in the early stages of vehicle development, can optimize in-vehicle speech clarity in the early stages of vehicle development, can greatly shorten the vehicle development cycle, and has high application value.

The embodiment of the invention provides a voice definition control method and device and electronic equipment, and the method comprises the steps that the electronic equipment obtains a to-be-processed first voice signal, determines the signal-to-noise ratio of the first voice signal, obtains a formant of a first acoustic signal, and obtains the formant of the first acoustic signal according to the signal-to-noise ratio of the first voice signal; determining an adjustment amount of a formant of the first acoustic signal; therefore, the definition of the voice signal heard by the user using the electronic equipment can be adjusted according to the current scene of the electronic equipment. And finally, the electronic equipment generates a second acoustic signal according to the adjustment amount, the upper bound and the lower bound of the formant of the first acoustic signal, the fundamental tone frequency of the first acoustic signal and the signal spectrum corresponding to the second voice signal, obtains a to-be-played third voice signal according to the second acoustic signal, and plays the third voice signal.

The invention provides a voice definition testing method, a voice definition testing device and a train. The method comprises the following steps: acquiring in-vehicle layout information and vehicle component setting parameters of a railway vehicle; determining a voice pickup area in the railway vehicle according to the in-vehicle layout information and the vehicle component setting parameters, so as to install sound acquisition equipment in the voice pickup area; generating a target excitation voice signal by using the test sound source under the operation condition that the rail vehicle is at the target vehicle speed, so as to acquire a voice pulse signal by using sound acquisition equipment; for any voice pulse signal, intercepting a target voice signal from the voice pulse signal based on the signal smoothness; and generating in-vehicle voice clear data of the railway vehicle according to the plurality of target voice signals of the different operation conditions.

The invention discloses a voltage-sharing non-inductive sound amplification and omnidirectional microphone integrated conference system, which relates to the technical field of conference audio, comprises a distributed loudspeaker array, an omnidirectional microphone array and a signal processing unit, and is characterized by further comprising a structure sound feedback suppression subsystem, a structure sound amplification subsystem and a structure sound amplification subsystem, the structural sound feedback suppression subsystem comprises a vibration sensing module, a self-adaptive filtering module, a signal synthesis module and a feedback analysis module; the vibration sensing module is rigidly installed inside the housing of the omnidirectional microphone unit or is directly coupled with the installation structure of the microphone array. According to the voltage-sharing non-inductive sound amplification and omnidirectional microphone integrated conference system, through a multi-stage cooperative active suppression mechanism, the problem of sound feedback caused by vibration conducted by a physical structure in a highly integrated conference system is effectively solved, the voice definition and the system sound amplification gain are improved, and therefore high-quality non-inductive sound amplification and pickup experience is achieved.

The present invention belongs to the field of speech signal acquisition technology, and specifically relates to a speech signal acquisition method and system. The method comprises: sampling and framing an analog signal to obtain a plurality of speech frames; obtaining the information density of each speech frame based on the short-time energy and zero-crossing rate of each speech frame, wherein the information density is positively correlated with the short-time energy and zero-crossing rate; constructing a quantization function for each speech frame based on the information density of each speech frame; and quantizing and encoding each speech frame based on the quantization function to obtain a digital speech signal. The present invention adapts the quantization function of each speech frame to its information density, thereby improving speech clarity and intelligibility.

The invention discloses a report hall sound reinforcement system based on an SIS system principle, and relates to the technical field of control systems, and the system comprises a multichannel input interface, a main sound reinforcement array, an auxiliary loudspeaker array, an ultra-low frequency loudspeaker, a matrix processor, and a self-adaptive controller. The directional radiation design of the main sound amplification array is combined with the distributed auxiliary loudspeaker topology, so that the whole-field sound pressure uniformity is improved, and the problem of non-uniform sound field distribution of a traditional system is solved; a middle sound channel independent regulation and control technology optimizes music signal restoration quality while guaranteeing language definition through a dynamic frequency division point switching mechanism, solves a long-term pain point of compatibility of music and human voice in the industry, effectively inhibits crosstalk between sound channels through a phase correction algorithm of a matrix type audio processing module, improves sound positioning precision, and improves sound positioning accuracy. A sound positioning fuzzy phenomenon is eliminated; the ground coupling design of the ultralow-frequency module reduces low-frequency harmonic distortion, realizes high-efficiency transmission of low-frequency energy and avoids interference on medium-high-frequency sound image positioning.

The invention relates to the field of voice signal processing and deep learning, and discloses a real-time full-band voice noise reduction enhancement method based on a hybrid DSP deep learning method, and the method comprises the following steps: signal preprocessing: carrying out the framing processing of an input time domain voice signal, obtaining a series of voice frames, and applying a window function to each frame of voice signal; frequency spectrum division: performing fast Fourier transform on each frame of signal to obtain frequency spectrum data, and dividing the frequency spectrum data into a plurality of frequency bands to form a multi-frequency-band frequency spectrum; and feature extraction: extracting features of each frequency band, and processing the features through a convolutional neural network. According to the method, the frequency band gain and the gain smoothness are optimized through the mixed DSP deep learning method, the voice definition and the voice intelligibility are improved, the robustness of the model in a complex noise environment is enhanced, the calculation complexity is reduced through multi-frequency band processing and algorithm optimization, low-delay and efficient real-time voice processing is ensured, and the method is suitable for low-power-consumption equipment.

A training method for a multimodal speech language large model is provided. The implementation is: obtaining first response speech data generated by the multimodal speech language large model by inputting first inquiry speech data into the multimodal speech language large model; determining an inquiry text corresponding to the first inquiry speech data and a response text corresponding to the first response speech data; determining, based on the inquiry text and the response text, a first score; determining, based on speech features of the first inquiry speech data and speech features of the first response speech data, a second score, where the speech features include at least one of speech clarity, speech rate feature, timbre feature, intonation feature, and emotion feature; and adjusting, based on the first score and the second score, parameters of the multimodal speech language large model.

The present invention provides a sound processing method, system, device and storage medium based on hearing aids, the method includes large-scale collection of sound signals of the hearing aid use environment, including background noise and speech signals; preprocessing the sound signals, processing outliers and missing values, and scaling them to a uniform scale, dividing them into training sets and test sets; using an improved gradient descent optimization algorithm to adjust and optimize the parameters of the standard adaptive filtering model, the parameters include filter order, maximum number of iterations, initial step size and regularization coefficient; inputting the test set into the optimized algorithm to achieve optimized processing of the sound signal. The present invention can improve speech clarity and background noise suppression capabilities, and improve the use effect and user experience of hearing aids in complex environments.

The present application discloses an audio processing method, a computer device and a computer-readable storage medium, which belong to the field of audio technology. The method comprises: obtaining a first audio signal collected by a first earphone, and obtaining a second audio signal collected by a second earphone. Afterwards, determining whether the ambient sound source includes a target sound source located in a target sound emission area based on the first audio signal and the second audio signal. If the ambient sound source includes the target sound source, speech enhancement processing is performed based on the first audio signal and the second audio signal to obtain a first target audio signal; if the ambient sound source does not include the target sound source, noise filtering processing is performed based on the first audio signal and the second audio signal to obtain a second target audio signal. In the present application, speech enhancement is performed on the sound generated by the target sound source in the target sound emission area, and noise filtering is performed on the sound generated by other sound sources outside the target sound emission area, which can improve the speech clarity of the target audio signal finally obtained.

The present application relates to the field of speech signal technology, and provides a speech enhancement method, device and electronic device. The method comprises: converting an input signal into the frequency domain, obtaining a first spectrum signal, and obtaining a smooth periodogram; using a first causal window to determine a first power spectrum level value corresponding to an energy spectrum value; using each first power spectrum level value to calculate a power spectrum density estimate of a steady-state component in the first spectrum signal; using a second causal window and a non-causal window to determine a second power spectrum level value as an energy spectrum value; using each second power spectrum level value to calculate a power spectrum density estimate of a non-steady-state component in the first spectrum signal; processing the first spectrum signal according to the power spectrum density estimate of the steady-state component and the non-steady-state component to obtain a speech output signal. The present application can effectively suppress non-steady-state noise and improve speech clarity by using a causal window and a non-causal window to determine the power spectrum density estimate of each component for the energy spectrum value.