31 results about "Masking threshold" patented technology

The invention belongs to the technical field of audio equipment design, and relates to a microphone for AI voice cloning based on pseudo tone of acoustic shielding, and the microphone comprises an audio collection module which is used for collecting an environment audio signal in real time; the analog-to-digital converter is used for converting the environment audio signal into a digital signal; the feature extraction module is used for performing frequency domain conversion on the digital signals and extracting frequency spectrum features and tone features; the acoustic shielding calculation module is used for calculating a shielding threshold value of the environment audio signal and determining a frequency band which cannot be heard by human ears in the environment audio signal according to the shielding threshold value; the pseudo tone generation module is used for generating pseudo tones based on the extracted tone features; the synthesis module is used for replacing a corresponding frequency band of the environment audio signal with the generated pseudo tone to form a synthesized audio; and the audio output module is used for outputting the synthesized audio. According to the invention, the pseudo tone can be generated in real time in the audio recording process, and it is ensured that a user is always protected during audio recording.

The invention discloses an audio watermark embedding and extracting method and system, and the method comprises the steps: obtaining audio data, calculating a masking threshold value of each frequency point of the audio data, and carrying out the preprocessing of the audio data; encoding according to the watermark information to generate an audio watermark, and embedding the audio watermark into the audio data according to a masking threshold value of the audio data; and retrieving the position of the audio watermark in the audio data embedded with the audio watermark, and extracting the audio watermark according to the masking threshold value of the original audio data. According to the method, the audio watermarks are embedded according to the masking theory of human ears, the masking threshold value of the audio data is calculated, and the audio watermarks are added to the audio according to the masking threshold value in the frequency domain, so that the audio watermarks can better resist attacks, the robustness of the audio watermarks is ensured, and the detectability can be kept even under severe conditions. The problem that the audio quality is easily affected due to insufficient robustness of the existing audio watermark is solved.

The invention provides an audio processing method, an audio processing device and a computer storage medium. The audio processing method is applied to the bathroom sound system. The audio processing method comprises the following steps: acquiring an original playing audio and a mixed audio signal acquired by a microphone; extracting a residual noise signal based on the original playing audio and the mixed audio signal; obtaining a dynamic gain based on the residual noise signal and a preset auditory masking threshold; and processing the original playing audio according to the dynamic gain, and obtaining and outputting a gain playing audio. Through the audio processing method, dynamic gain compensation can be realized for a bathroom scene, audio output can be dynamically adjusted according to environmental noise, intelligent noise reduction and adaptive adjustment are realized, and a bathroom sound box is ensured to provide rich and fine sound effect performance.

The application provides an audio processing method, an audio processing device and a computer storage medium. The audio processing method is applied to a bathroom sound system. The audio processing method comprises the following steps: obtaining original playing audio and mixed audio signals collected through a microphone; extracting a residual noise signal based on the original playing audio and the mixed audio signals; obtaining a dynamic gain based on the residual noise signal and a preset auditory masking threshold; processing the original playing audio according to the dynamic gain, and obtaining and outputting gain playing audio. Through the audio processing method, dynamic gain compensation can be realized for the bathroom scene, the audio output can be dynamically adjusted according to environmental noise, intelligent noise reduction and adaptive adjustment can be realized, and it is ensured that the bathroom sound system can provide rich and delicate sound effect performance.

This application discloses a method and system for dynamic mapping of partial discharge audio signals based on multimodal intelligent sensing, relating to the field of signal processing. The method includes: acquiring ambient audio signals and partial discharge signals; performing frequency domain conversion on the ambient audio signals to obtain a frequency domain energy distribution; calculating the masking sound pressure level based on the frequency domain energy distribution and constructing a global masking threshold curve; traversing the global masking threshold curve to extract the center frequency of the frequency band, and determining the center frequency of the frequency band with the lowest value as the dynamic carrier frequency; extracting the discharge pulse amplitude and pulse rise time characteristics from the partial discharge signal; mapping the inverse ratio of the pulse rise time characteristics to the modulation frequency, and mapping the direct ratio of the discharge pulse amplitude to the frequency modulation index; generating a synthesized audio stream; and outputting the synthesized audio stream to external headphones. This application can effectively sense and identify the converted partial discharge audio signal in complex noise backgrounds.

The invention relates to the technical field of audio signal processing, and discloses an audio processing method and system based on auditory model optimization. The method comprises the following steps: performing multi-scale frequency band decomposition processing on an input audio signal to generate a sub-band signal set covering different frequency ranges; the sub-band signal set is input into an auditory masking effect calculation model, and the model calculates masking threshold distribution of each sub-band signal according to frequency sensitivity characteristics of a human auditory system. And performing dynamic gain adjustment processing on the sub-band signal set based on the calculated masking threshold distribution to generate a gain-optimized sub-band signal set. And performing nonlinear distortion suppression processing on the sub-band signal set after gain optimization to generate a sub-band signal set after distortion suppression. And inputting the sub-band signal set after distortion suppression into a frequency band synthesis model to reconstruct a complete time domain audio signal. According to the invention, audio optimization processing conforming to human ear perception characteristics is realized.

The invention relates to the technical field of audio coding, and particularly discloses an audio coding method and device, equipment, a medium and a product, and the method comprises the steps: carrying out the spectrum analysis of an input audio frame, and judging whether a single-frequency signal meeting a preset condition exists or not; if yes, determining that a scale factor band where the single-frequency signal is located and N scale factor bands adjacent to the scale factor band form a target scale factor band combination; setting the masking threshold value of each scale factor band in the target combination to be smaller than the initial masking threshold value, and setting the masking threshold value of each scale factor band in the non-target combination to be larger than the initial masking threshold value; and selecting a scaling factor based on the masking threshold of each scaling factor band, and quantifying each scaling factor band according to the scaling factor to obtain a coded bit stream. Therefore, under the condition that the bit budget is limited, the coding fidelity of the single-frequency signal is improved, and the objective distortion index (for example, the objective distortion index of THD + N is improved) is improved.

The invention belongs to the technical field of audio data processing, and relates to a pseudo tone anti-AI voice cloning method and system based on acoustic shielding, and the method comprises the steps: S1, constructing a pseudo tone based on the tone characteristics of a given audio file material; s2, determining a masking threshold value of the target audio according to the intensity and the spectrum characteristics of the target audio, and marking the frequency band with the power lower than the masking threshold value as an unaudible frequency band; and S3, embedding the pseudo tone into an unaudible frequency band of a target audio to obtain a synthesized audio. According to the method and the device, active defense of AI voice cloning can be realized on the premise of not influencing the tone quality, relatively high universality and robustness are achieved, and various voice cloning attacks can be resisted.

Embodiments are related to a masking threshold determinator, wherein the masking threshold determinator is configured to obtain a plurality of bandpass signals using a plurality of filters having different bandwidths; and wherein the masking threshold determinator is configured to obtain a masking threshold information associated with a given frequency region on the basis of bandpass signal values of at least two bandpass signals.Furthermore, audio encoders, methods and computer programs are disclosed.

The present invention discloses an active noise control system and method based on a convolutional recurrent neural network. The system consists of a preprocessing module, a convolutional recurrent neural network, a signal reconstruction module, and a network loss module. The preprocessing module is used to preprocess a reference signal and a sound signal and extract corresponding features. The convolutional recurrent neural network consists of two convolution modules, a recurrent neural network, and two deconvolution modules. The reference signal features and the sound signal features are used as network inputs to predict the elimination signal features. The signal reconstruction module is used to reconstruct the elimination signal features in the frequency domain into an elimination signal in the time domain. The elimination signal is played through a loudspeaker to achieve the purpose of noise control. The network loss module consists of a psychoacoustic model and a loss calculation module. The masking threshold is obtained through the psychoacoustic model, and the total network loss is calculated. The gradient is back-propagated and used to update the weights of the network module.

This disclosure relates to information processing apparatuses and methods that can suppress both the reduction in subjective quality of reproduced sound and the reduction in coding efficiency. In this disclosure, a direction-defined bitstream is generated by encoding sound data using a bit allocation based on a direction-defined spatial masking threshold, where the direction-defined spatial masking threshold is a spatial masking threshold that defines the directional range of a sound source. The direction-defined bitstream generated by encoding sound data using a bit allocation based on the direction-defined spatial masking threshold is acquired, and the acquired direction-defined bitstream is decoded, where the direction-defined spatial masking threshold is a spatial masking threshold that defines the directional range of a sound source. This disclosure can be applied, for example, to information processing apparatuses or information processing methods.

This invention relates to the field of audio signal processing technology, and discloses a method and system for dynamic audio control optimization based on convolutional neural networks. The method includes: separating the mixed audio signals using a source separation convolutional neural network to obtain independent estimated spectra for each source category; extracting energy demand prediction vectors and transient feature descriptors for each source category using a spectral residual convolutional neural network; calculating the driving characteristic matching score between each source category and each speaker unit; performing propagation calculations based on the room transfer function and constructing an arrival power prediction matrix; calculating the masking threshold matrix and perceived loudness estimate for each listener position; solving for the optimal power allocation coefficient through a joint optimization objective function; extracting position-specific perception sensitivity coefficients using an auditory masking perception convolutional neural network; generating adaptive dynamic compression gain values ​​and finally outputting the driving signals for each speaker unit.

A masking threshold determinator for providing a masking threshold information on the basis of an input audio signal is configured to obtain envelope information describing envelopes of different frequency ranges of the input audio signal. The masking threshold determinator is configured to determine a comodulation strength information describing a comodulation between different frequency ranges of the input audio signal. The masking threshold determinator is configured to apply a post-masking modeling to the envelope information, or to a pre-processed version thereof, to obtain a post-masking-processed envelope information, which describes a masking threshold. The masking threshold determinator is configured to adapt a post-masking decay time of the post-masking modeling in dependence on the comodulation strength information. A quantization step size determinator and an encoder, corresponding methods and a computer program are also disclosed.

The invention relates to an audio signal compression processing method and a related device thereof. The method comprises the following steps: acquiring feature information of an audio signal to be compressed; wherein the feature information at least comprises time domain information, frequency domain information and classification information; inputting the feature information of the audio signal into a preset psychological acoustic model to obtain a perception masking threshold value of the audio signal, and determining a compression parameter of the audio signal based on the feature information and the perception masking threshold value; and based on the compression parameter, compressing the audio signal to be compressed, and generating a compressed target audio signal. According to the scheme provided by the invention, the compression parameter can be dynamically adjusted according to the audio signal, redundant information is reduced, and the compression efficiency is improved while key audio information perceived by human ears is reserved.

Techniques are described for hybrid masking threshold-based perceptual slacks for audio watermarking. In some embodiments, the techniques include identifying an audio signal, determining perceptual slacks for the audio signal, generating a watermarked audio signal that includes an audio watermark based on the perceptual slacks, and outputting the watermarked audio signal using one or more speakers, for localization of the one or more speakers.

The invention discloses an engine power adjusting method, device and equipment, a readable storage medium and a vehicle. The method comprises the steps that an initial audio signal in a vehicle cabin is collected; filtering a first noise signal in the initial audio signal to obtain a target audio signal, the first noise signal being a noise signal emitted by an engine of the vehicle; the target audio signal is used as masking sound for masking effect analysis, a masking threshold value for a noise signal emitted by the engine is obtained, and the masking threshold value represents the minimum sound intensity capable of being sensed by human ears; the power of the engine is adjusted based on the masking threshold. According to the method, the masking effect of the environmental noise on the noise generated by the engine during working is considered, so that the power of the engine is adjusted based on the initial audio signal, the influence of the noise generated by the engine on a user in the vehicle cabin can be reduced, and the noise reduction effect in the vehicle cabin is improved.

The application belongs to the technical field of audio device design, and relates to a microphone for pseudo-tone color confrontation AI voice cloning based on acoustic masking, which comprises an audio acquisition module, an analog-to-digital converter, a feature extraction module, an acoustic masking calculation module, a pseudo-tone color generation module, a synthesis module and an audio output module. The audio acquisition module is used for acquiring an environmental audio signal in real time. The analog-to-digital converter is used for converting the environmental audio signal into a digital signal. The feature extraction module is used for performing frequency domain conversion on the digital signal, and extracting frequency spectrum features and tone features. The acoustic masking calculation module is used for calculating a masking threshold of the environmental audio signal, and determining a frequency band that is inaudible to human ears in the environmental audio signal according to the masking threshold. The pseudo-tone color generation module is used for generating a pseudo-tone color based on the extracted tone features. The synthesis module is used for replacing the corresponding frequency band of the environmental audio signal with the generated pseudo-tone color to form a synthesized audio. The audio output module is used for outputting the synthesized audio externally. The application can generate a pseudo-tone color in real time during audio recording, and ensures that a user is always protected during audio recording.

The embodiment of the disclosure discloses a psychoacoustic analysis method, device, equipment and storage medium, which can be applied to a communication system. The method comprises the following steps: determining a plurality of masking sources of an audio signal; and analyzing a masking threshold of the audio signal according to part of the masking sources. By implementing the method of the disclosure, the calculation amount of psychoacoustic analysis can be effectively reduced, and the calculation complexity can be reduced by selecting part of the masking sources from all the masking sources of the audio signal to participate in the analysis and calculation of the masking threshold.

This invention relates to the field of teaching evaluation technology, specifically to a hybrid teaching evaluation system based on artificial intelligence. The system includes an audiovisual coupling tagging module, a frequency domain fingerprint embedding module, a gaze vector gating module, an interaction intent verification module, and a teaching effectiveness rating module. In this invention, by extracting the pixel displacement modulus of the blackboard and cross-correlation with the audio energy envelope, the system accurately segments substantive teaching segments using physical signal synchronicity, automatically eliminates invalid silences to focus on high-value intervals, embeds markers at high frequencies in the audio using auditory masking thresholds, binds sound waves to the real space to prevent the absence of online users, constructs a three-dimensional normal vector angle model between the face and the screen to determine orthogonal focusing of gaze, combines screen displacement logic verification to eliminate false data related to mechanical gaze, calculates the ratio of effective cognitive injection time to teaching segments, quantifies net teaching effectiveness, and accurately maps the actual conversion efficiency of knowledge transfer.

Disclosed are an audio signal processing method using a psychoacoustic model and an electronic device for performing same. The audio signal processing method may comprise: an operation of receiving an input audio signal; an operation of converting the input audio signal into an audio signal in the frequency domain; an operation of dividing the audio signal in the frequency domain into audio signals of a plurality of subbands; an operation of determining a masking threshold corresponding to each of the subbands by using a psychoacoustic model that takes the audio signals of the subbands as inputs; an operation of acquiring audio vector data corresponding to each of the subbands by using neural network–based encoders; and an operation of generating a compressed audio signal by quantizing the audio vector data corresponding to each of the subbands on the basis of the masking threshold corresponding to each of the subbands.

Disclosed in the present invention are an audio watermark embedding and extraction method and system. The method comprises: acquiring audio data, calculating a masking threshold for each frequency point of the audio data, and preprocessing the audio data; performing encoding on the basis of watermark information to generate an audio watermark, and embedding the audio watermark into the audio data on the basis of the masking threshold of the audio data; and locating the position of the audio watermark in the audio data into which the audio watermark is embedded, and extracting the audio watermark on the basis of the masking threshold of the original audio data. In the present invention, an audio watermark is embedded on the basis of human auditory masking theory; the masking threshold of audio data is calculated, and the audio watermark is added to audio in a frequency domain on the basis of the masking threshold, thereby allowing the audio watermark to better resist these attacks, ensuring the robustness of the audio watermark, and ensuring that the detectability can be maintained even under harsh conditions. The present invention solves the problem whereby existing audio watermarks have insufficient robustness and are prone to affect the quality of audio itself.

The invention discloses a power control method, device and equipment, a readable storage medium and a vehicle, and the method comprises the steps: filtering a signal generated by the work of each device of the vehicle from a sensing signal collected by a vehicle sensor, and obtaining a background noise signal; the background noise signal is used as masking sound for masking effect analysis, a first masking threshold value for an audio signal emitted by the target device is obtained, and the first masking threshold value represents the minimum sound intensity capable of being sensed by human ears; and adjusting the power of the target device according to the first masking threshold. According to the method, the power of the target device is adjusted through masking effect analysis, the power of the target device can be continuously adjusted according to the change of the environmental noise, the influence of the noise emitted by the target device on a driver and passengers is reduced, and meanwhile, due to the fact that the adjustment of the power of the target device is carried out based on the actual situation, the driving safety is improved. Energy distribution can be more flexible, and excessive energy loss is avoided.

This application discloses a psychoacoustic speech masking method for proactive defense against speech cloning, relating to the field of speech information security. The method includes: performing time-frequency decomposition on the original reference speech using time-frequency analysis to obtain speech information in the speech representation space; adding a look-ahead perturbation to the speech information to obtain look-ahead speech; determining a speaker confusion loss using a speaker coding network based on the original reference speech and the look-ahead speech; determining a total masking threshold using a psychoacoustic masking model based on the speech information in the speech representation space; determining a speech quality constraint loss based on the perturbation power spectrum according to the total masking threshold; determining a total loss based on the speaker confusion loss and the speech quality constraint loss; generating a final perturbation using an iterative optimization algorithm based on the total loss; and generating a final protected reference speech based on the final perturbation and the original reference speech. This application weakens the ability to extract the identity of the real speaker, thereby reducing the risk of speech cloning and voiceprint impersonation.

The invention discloses an audio compression model processing method and system, equipment and a program product, and the method comprises the steps: compressing a sample audio through an initial audio compression model, and obtaining a predicted audio; respectively dividing the sample audio and the predicted audio into a sensitive frequency band and a non-sensitive frequency band; dividing sensitive frequency bands of the sample audio and the predicted audio into four frequency bands based on a first auditory masking threshold value, and determining a weight corresponding to a frequency spectrum square difference value of the frequency bands in the same range of the sample audio and the predicted audio, dividing the non-sensitive evaluation of the sample audio and the predicted audio into four frequency bands based on a second auditory masking threshold, and determining a weight corresponding to a frequency spectrum square difference value of the frequency bands of the same range of the sample audio and the predicted audio; obtaining the frequency spectrum Mel sub-band loss by using each frequency spectrum square difference value and the corresponding weight; and optimizing the initial audio compression model based on the spectrum Mel sub-band loss to obtain a target audio compression model.

The present application relates to a call audio processing method, device, computer equipment and computer-readable storage medium. The method includes: obtaining call audio sent by multiple call member terminals participating in a multi-person call; selecting one of the call member terminals participating in the multi-person call as a receiving terminal, and obtaining the call environment noise of the receiving terminal; determining the masking degree of each call audio according to the call environment noise and the masking threshold of each call audio; the masking degree indicates the degree to which the call audio is masked by the call environment noise; and filtering the call audio according to each masking degree and sending it to the receiving terminal. This method can improve the call quality of multi-person calls.

This invention discloses an audio feature extraction method based on a perceptual-semantic dual-domain approach, comprising the following steps: Step 1: Obtaining an audio signal A and its critical frequency band feature B; Step 2: Calculating the simultaneous masking effect of the audio signal A, characterized by a joint masking threshold. Spectral components below the joint masking value will not be perceived by the human ear. After calculating the joint masking threshold, the perceptual importance of the audio signal A is determined by comparing the joint masking threshold with the actual energy of the audio signal A; Step 3: Inputting the critical frequency band feature B into an audio semantic encoder, and projecting the audio features of the audio signal A, which incorporates type labels and emotional attributes, into a 256-dimensional compact semantic space through a semantic embedding generation network to generate a semantic embedding vector. This invention effectively obtains both the perceptual importance and semantic embedding vector features of audio.

This invention relates to the field of fan, speaker, and light design and control technology, and discloses a low-noise design method for a speaker fan based on fan noise. The method includes receiving the original audio signal, generating an audio power spectrum, and extracting low-frequency transient features; calculating the instantaneous masking threshold using a psychoacoustic model, and then inversely shaping the allowable noise power spectrum of the fan motor to generate a target fan spectrum; generating a fan drive signal based on the target spectrum using a random spread spectrum algorithm, and constructing an airflow velocity vector by combining the angular velocity fed back from the motor and the duct parameters; calculating the phase compensation amount based on the airflow velocity vector to generate a phase correction signal, and then combining the low-frequency transient features with an anti-phase aerodynamic impedance compensation component to synthesize a speaker signal; finally, driving the fan motor and speaker unit to operate synchronously. This invention utilizes the auditory masking effect to achieve invisible fan operation and cancels the interference of airflow on sound waves through flow field mapping and phase pre-distortion, achieving synergistic optimization of the sound field and airflow field.

The present disclosure pertains to an information processing device and method that make it possible to suppress reduction in coding efficiency while suppressing reduction in the subjective quality of reproduced sound. In the present disclosure, a direction-limited bit stream is generated by coding sound data with bit allocation based on a direction-limited spatial masking threshold that is a spatial masking threshold at which a sound source direction range is limited. The direction-limited bit stream generated by coding the sound data with bit allocation based on the direction-limited spatial masking threshold that is the spatial masking threshold at which the sound source direction range is limited is acquired, and the acquired direction-limited bit stream is decoded. The present disclosure can be applied, for example, to an information processing device or an information processing method.

The invention discloses an audio processing method and device and a storage medium. Wherein the original audio data is subjected to equal-length slicing to obtain a plurality of audio slices, a masking threshold curve corresponding to the current audio slice is determined, and the masking threshold curve is used for indicating a first masking threshold corresponding to the frequency index of each masker; playing the current audio slice to the user, collecting an electroencephalogram signal of the user, and determining an electroencephalogram signal slice corresponding to the current audio slice; and according to the spectrum characteristics of the gamma frequency band in the electroencephalogram signal slice and the masking threshold curve, enhancing the energy of the gamma frequency band in the next audio slice of the current audio slice to obtain the enhanced next audio slice, so that enhancement of the energy of the gamma frequency band in the audio slices and auditory experience of a user can be considered; therefore, the treatment effect is improved.