7 results about "A-weighting" patented technology

The invention discloses a multi-channel active noise control method capable of reducing the calculated amount. The method comprises the following steps: (1) building a multi-channel ANC system, wherein the system comprises M (Mgt; 1) the number of the microphones; (2) measuring a secondary channel transfer function; (3) selecting Q (Qlt; m) microphones, calculating a weighting matrix by using the latest T signals collected by the microphone system, and weighting the signals of the selected Q microphones to obtain a signal of a Q channel for controlling the updating of a filter; (4) updating a control filter; and (5) calculating the output of the control filter, and inputting the output into the secondary sound source for playing. According to the method, the Q microphones are randomly selected in each period of time, and the Q microphone signals are weighted to obtain the update signal of the control filter, so that the calculation amount can be greatly reduced, and the global noise reduction effect can be obtained.

The application particularly relates to a target direction estimation method based on weighted deconvolution, which comprises the following steps: receiving far-field signals of a plurality of sound sources by using a sensor array, and constructing array receiving signals according to the far-field signals received by sensors at different positions in the array; configuring an array receiving signal data matrix of a multi-shot condition based on the array receiving signals; determining a covariance matrix corresponding to the array receiving signal data matrix; configuring a weighting vector of a conventional beamformer, and scanning the covariance matrix in different directions in a two-dimensional space by using the conventional beamformer to obtain a corresponding spatial spectrum; calculating the spatial spectrum based on the relationship between a sound source intensity distribution function and a point spread function to obtain a point spread function corresponding to the spatial spectrum; and calling an improved R-L method to perform direction estimation according to the point spread function and the spatial spectrum, and determining the position information of the sound source. The method can be applied to sensor arrays of any shape, and the resolution capability for multiple targets is improved.

The application relates to a sound signal denoising method, device and equipment and a storage medium, which can be used in the technical field of signal processing. The method comprises the following steps: performing variational mode decomposition processing on a to-be-denoised sound signal to obtain a plurality of intrinsic mode components; determining a weighting coefficient of each intrinsic mode component according to a mutual information value between each intrinsic mode component and the to-be-denoised sound signal; performing weighting reconstruction processing on each intrinsic mode component according to the weighting coefficient to obtain a reconstructed signal corresponding to the to-be-denoised sound signal; performing updating processing on wavelet coefficients obtained after multi-scale wavelet decomposition of the reconstructed signal to obtain target wavelet coefficients; and performing inverse wavelet transform processing on the target wavelet coefficients to obtain a denoised sound signal corresponding to the to-be-denoised sound signal. The method can improve the accuracy of sound signal denoising.

A method for equalizing an audio frequency signal broadcast in a broadcasting environment. Such a method includes: estimating a frequency profile of a noise signal representing a background noise present in the broadcasting environment, based on, on the one hand, a signal captured by at least one microphone implemented in the broadcasting environment and, on the other hand, the audio frequency signal; determining a desired frequency profile for the broadcast audio frequency signal; determining an acoustic frequency mask representing, for each frequency component, a difference between the frequency profile of the noise signal and the desired frequency profile; and equalizing the audio frequency signal via a weighting of a spectrum of the audio frequency signal by applying a frequency weighting mask that is a function of the acoustic frequency mask, delivering the equalized audio frequency signal.

The present disclosure provides an audio processing method, an audio processing apparatus and device, and a computer-readable storage medium. The audio processing method includes: acquiring a set of audio data; evaluating each audio data by using a plurality of objective metrics to generate an objective metric vector; acquiring a correlation vector representing correlation between the plurality of objective metrics and a plurality of subjective metrics, and determining a weighting coefficient vector for each audio data based at least on the correlation vector; performing weighted summation on the objective metric vector by using the weighting coefficient vector to obtain an evaluation score; and screening, from the set of audio data, audio data with the evaluation score greater than a first predetermined threshold into a set of target audio data. The set of target audio data is used for establishing training data pair for training neural networks.

The invention discloses a near-field acoustical holography optimization method based on a compressed sensing equivalent source. The near-field acoustical holography optimization method comprises the following steps of 1) performing equivalent processing on a space sound field; step 2), selecting the position and size of a holographic measurement surface in a potential sound source near field, performing equidistant grid division on the holographic surface, and determining a sampling strategy; 3) selecting a certain number of measurement points on the holographic measurement surface in a random manner, constructing a near-field acoustic holographic model based on the compressed sensing equivalent source, and solving an initial sparse solution of the source intensity of the equivalent source by using L1 regularization according to the sound pressure at the measurement points randomly selected in the holographic surface; according to the method, the sparsity of an equivalent source intensity solution can be further improved, the influence of a pseudo source is effectively restrained, the convergence speed of iterative calculation can be increased, and the calculation efficiency is improved.

The invention discloses a real-time audio analysis and diagnosis method for the inspection process of a continuous annealing furnace. The method comprises the following steps: S1, waiting for an audio file address of rotary equipment to be transmitted into a text file of an audio file storage path; s2, reading an audio file address; s3, acquiring audio file information; s4, performing A weighting on the audio data in the audio file; s5, calculating the total sound pressure level decibel of the audio data weighted by A in each time period; s6, carrying out short-time Fourier time-frequency analysis on the audio data weighted by A to obtain amplitudes at all moments and frequencies; s7, calculating 1 / 3 octave of the audio data weighted by A, and obtaining a sound pressure level mean square value in each octave; s8, storing analysis and diagnosis results of the steps S5-S7; and S9, emptying the text file for storing the audio file path. According to the invention, the sound data is analyzed and diagnosed in real time, the diagnosis result is stored in real time, and the real-time performance and automation of furnace area equipment monitoring are realized.