34results about How to "Improve hearing quality" patented technology

The hearing comfort when wearing hearing apparatuses is to be improved. To this end, a hearing apparatus and in particular a hearing device is proposed, which has a directional microphone with a preferred direction of recording sound. Furthermore, the hearing apparatus has an inclination determination facility, for determining an inclination of the preferred direction of recording sound in respect of the absolute horizontal for instance. An alignment unit then provides for the directional microphone to be aligned as a function of the determined inclination. The directional microphone of a hearing device can thus always be aligned horizontally independently of the movement of the head of a hearing device wearer.

The invention discloses a tone decoding method and a device. The method comprises the following steps: utilizing the low-frequency band spectrum of the current frame of a narrow-frequency band to generate the high-frequency band spectrum of the current frame; judging whether the narrow band frame needs to be amended under certain conditions, and amending the high-frequency band spectrum of the narrow band frame in needing amending; changing the current frame which is treated according to a judging result from a frequency domain signal to a time domain signal; utilizing the attenuation factor of the current frame to attenuate the high-frequency time domain signal of the time domain signal of the current frame, and then outputting the time domain signal of the current frame. Correspondingly, the invention also discloses the tone decoding device, comprising an expansion unit, a first amending processing unit, a changing unit and an attenuation unit. By the tone decoding method and the device, the embodiment of the invention greatly solves the problem of uncomfortable hearing caused by the tone signals of different bandwidths and improves the hearing quality of the tone signals.

The invention relates to a method for eliminating indoor reverberation and belongs to the technical field of signal processing. A later reverberation power spectrum estimation module, a spectrum subtraction module, a sound / sound-free detection module, an energy attenuation module and an adaptive spectrum line enhancement module are used in the method; input of the later reverberation power spectrum estimation module is a reverberation voice, and output of the later reverberation power spectrum estimation module is connected with the spectrum subtraction module; input of the spectrum subtraction module is the reverberation voice and the output of the later reverberation power spectrum estimation module, and output of the spectrum subtraction module is connected with the sound / sound-free detection module; output of the sound / sound-free detection module controls the output of the spectrum subtraction module to be selectively connected with the energy attenuation module or the adaptive spectrum line enhancement module; the energy attenuation module and the adaptive spectrum line enhancement module output a final enhanced voice.

The invention discloses a prior signal-to-noise ratio estimating method based on an MMSE error criterion and used for voice enhancement, and belongs to the technical field of voice signal processing. Aimed at the prior signal-to-noise ratio estimating problem in the voice enhancement technology, the method comprises the steps of: firstly carrying out preliminary estimation on a prior signal-to-noise ratio of noised voices based on the MMSE error criterion, carrying out Wiener filtering calculation on an obtained prior signal-to-noise ratio estimated value to obtain a first system gain factor, carrying out calculation on the first system gain factor an amplitude spectrum value of the noised voices to obtain a voice power spectrum estimated value, then utilizing the obtained voice power spectrum estimated value and a power spectrum estimated value of noise to carry out estimation once again, and obtaining a final prior signal-to-noise ratio estimated value. The prior signal-to-noise ratio estimated value is substituted into a subsequent voice enhancing step for processing, and de-noised estimated voice clearing signals are obtained. The prior signal-to-noise ratio estimating method based on the MMSE error criterion can effectively inhibit background noise components in estimated cleared voices, and excessive damages to the cleared voice components are avoided, so that the hearing quality of the estimated cleared voices is improved, and the performance of a voice enhancement algorithm is improved.

An active grille shutter assembly includes a frame adapted to be connected to a portion of a vehicle and forming at least one opening therein and a plurality of shutters disposed in and extending across the at least one opening and being pivotally connected to the frame. The active grille shutter assembly also includes at least one linkage connected to the shutters and an actuator supported by the frame and being connected to one of the shutters to move the at least one linkage to open and close the shutters as a unit.

The invention discloses a method and a device for processing a voice signal with noise, and a server, which belong to the technical field of communication. The method comprises the steps of acquiring a noise signal in the voice signal with noise according to a silence section of the voice signal with noise; for each frame in the voice signal, acquiring the power spectrum iteration factor of each frame in the voice signal according to the noise signal and the voice signal with noise; calculating the intermediate power spectrum of each frame according to the voice signal with noise, and the power spectrum iteration factors of each frame and the previous frame; calculating the signal to noise ratio of each frame in the voice signal with noise according to the intermediate power spectrum of each frame of the voice signal and the noise signal; acquiring the processed time-domain voice signal with noise according to the signal to noise ratio of each frame in the voice signal with noise, the voice signal with noise and each frame of the noise signal. Through processing the voice signal with noise through the power spectrum iteration factors, the hearing quality of a user is improved.

Provided in embodiments of the present invention are a prediction method and decoding device for a bandwidth expansion frequency band signal. The method comprises: demultiplexing a received bitstream to acquire a frequency domain signal; determining whether or not the highest frequency point having bits allocated of the frequency domain signal is less than a predetermined starting frequency of a bandwidth expansion frequency band; if yes, predicting an excitation signal of the bandwidth expansion frequency band on the basis of excitation signals in a predetermined frequency band range of the frequency domain signal and of the predetermined starting frequency point of the bandwidth expansion frequency band; otherwise, predicting the excitation signal of the bandwidth expansion frequency band on the basis of the excitation signals in the predetermined frequency band range of the frequency domain signal, of the predetermined starting frequency point of the bandwidth expansion frequency band, and of the highest frequency point having bits allocated; and, predicting a bandwidth expansion frequency band signal on the basis of the predicted excitation signal of the bandwidth expansion frequency band and of a frequency domain envelope of the bandwidth expansion frequency band. The technical solution of embodiments of the present invention is capable of effectively ensuring between preceding and subsequent frames the continuity of the excitation signals predicted for the bandwidth expansion frequency band signal, thus ensuring the auditory quality of a restored bandwidth expansion frequency band signal.

Push button insertion tool is a device used to help clean and install a receiver-in-the-canal hearing instrument into the ear. This product may be in the shape of a tapered tube, resembling a writing pen. The far end of the tube may be shaped to fit receivers of various sizes, and may include a slot on the side through which the microphone wire may extend. On the other end of the tube may be a push button. When pushed, it moves a rod through the tube to a push plunger, which inserts the receiver out of the tube and into the ear canal. A return spring and guide pin on the far end is designed to retract the push rod back to starting position.

The invention discloses a bridge connection computing method of a digital teleconference, which is characterized in that: a time delay vibrating processing mechanism is adopted for modifying a multi-sectional code stream; VAD voice activity detection which combines single frame detection with long time window detection and sample rate matching algorithm are used for reducing the invalid number of channels which enter bridge connection algorithm and reducing computing false rate; and short-term amplitude computing and funnel audio mixing computing are used for reducing operand. The invention has the benefit effects as follows: 1) the method adapts to large time delay vibrating under the condition of IPNET, can offer multi-sectional code stream modification, offer continuous and homogeneous voice code stream for terminals, and improve the audio quality of the decoded voice; 2) the adoption of the VAD voice activity detection and the sample rate matching algorithm can reduce the invalid number of channels which enter bridge connection algorithm and reduce computing false rate of the bridge connection; and 3) the adoption of short-term amplitude computing method and funnel mixing computing method can greatly reduce the operand, avoid bridge connection misjudgment caused by innovation shocks and improve the audio mixing quality of the bridge connection.

The embodiments of the invention disclose a method and apparatus for recovering noise components of audio signals, and provide an audio signal processing method and apparatus. The method comprises the following steps: receiving a code stream, and decoding the code stream to obtain the audio signals; according to the audio signals, determining first audio signals; determining a symbol of each sampling value in the first audio signals and an amplitude value of each sampling value; determining an adaptive normalization length; according to the adaptive normalization length and the amplitude value of each sampling value, determining an adjustment amplitude value of each sampling value; and according to the symbol of each sampling value and the adjustment amplitude value of each sampling value, determining the second audio signals. According to the embodiments of the invention, for the audio signals with rising edges or falling edges, when the noise components are recovered, echoes are not caused to the signals after the noise components of the audio signals are recovered, and the hearing quality of the signals after the noise components of the audio signals are recovered is improved.

The present invention discloses a digital audio camouflage and reconstruction method based on segmented sequences. The method comprises the steps of taking sub segment sequences obtained by dividing a published audio and a secret audio as a secret sequence and a published sequence, adding random perturbations to the sequences with the same element value, directly carrying out least squares matching on an equidistant transformation sequence obtained by the rotate right of the secret sequence and a corresponding position published sequence, finding a rotate right step length with a minimum residual error and a matched parameter, and thus camouflaging the secret audio as the published audio and further reconstructing the camouflaged audio as the secret audio. The method is easy to realize, since only the corresponding position equidistant transformation sequence matching is carried out, while the encoding time is reduced, the equidistant transformation number is raised, the matching precision is improved, the detail loss and overflow problems brought by the deformation from a smooth block to a complex block and the restoration from the smooth block to a complex texture block can be effectively avoided, and thus the auditory quality of camouflaged and reconstructed audios is improved further.

The invention provides a bone conduction speech enhancement method based on differential operation and joint dictionary learning. In the training stage, in an indoor noise-free environment, a double-microphone array composed of bone conduction microphones and air conduction microphones is used for synchronously collecting training voices; short-time Fourier transform is performed on training signals of the bone conduction speech and the air conduction speech to obtain time-frequency spectrum amplitudes, and differential time-frequency spectrum amplitudes of the time-frequency spectrum amplitudes are calculated; and a joint speech dictionary of the bone conduction speech time-frequency spectrum amplitude and the differential time-frequency spectrum amplitude is learned on the time-frequencyspectrum. And at a detection stage, short-time Fourier transform is performed on the bone conduction speech to obtain a time-frequency spectrum amplitude and a phase, the is projected amplitude on abone conduction speech sub-dictionary of the joint speech dictionary, and a differential speech time-frequency spectrum amplitude is reconstructed by using an obtained optimal sparse representation coefficient and a differential time-frequency spectrum amplitude sub-dictionary of the joint speech dictionary. A bone conduction voice time-frequency spectrum is compensated and finally short-time inverse Fourier transform is performed to obtain an enhanced bone conduction voice time-domain signal.

The invention provides an encoding method of a multi-sound-track signal and an encoder. The encoding method comprises the steps of obtaining the multi-sound-track signal of a current frame; determining an initial ITD value of the current frame; controlling the quantity of target frames permitted to continuously occur according to characteristic information of the multi-sound-track signal, whereinthe characteristic information comprises at least one of signal-to-noise ratio parameters of the multi-sound-track signal and peak features of the mutual relation number of the multi-sound-track signal, and the ITD values of the target frames reuse the ITD value of the previous frames of the target frames; determining the ITD value of the current value according to the ITD value of the current frame and the quantity of the target frames permitted to continuously occur; encoding the multi-sound-track signal according to the ITD value of the current frame. The encoding quality of the multi-sound-track signal can be improved.

A prediction method and a coding / decoding device for a high frequency band signal. The method comprises: acquiring a signal type of an audio signal and a low frequency band signal (100); the audio signal comprises the low frequency band signal and a high frequency band signal; acquiring a frequency domain envelope of the high frequency band signal on the basis of the signal type (101); predicting an excitation signal of the high frequency band signal on the basis of the low frequency band signal (102); and, restoring the high frequency band signal on the basis of the frequency domain envelope of the high frequency band signal and of the excitation signal of the high frequency band signal. The method and device allow for effective reduction in errors found between the high frequency band signal acquired by prediction and an actual high frequency band signal, and increases the accuracy of the predicted high frequency band signal.

The invention discloses a tone decoding method and a device. The method comprises the following steps: utilizing the low-frequency band spectrum of the current frame of a narrow-frequency band to generate the high-frequency band spectrum of the current frame; judging whether the narrow band frame needs to be amended under certain conditions, and amending the high-frequency band spectrum of the narrow band frame in needing amending; changing the current frame which is treated according to a judging result from a frequency domain signal to a time domain signal; utilizing the attenuation factor of the current frame to attenuate the high-frequency time domain signal of the time domain signal of the current frame, and then outputting the time domain signal of the current frame. Correspondingly, the invention also discloses the tone decoding device, comprising an expansion unit, a first amending processing unit, a changing unit and an attenuation unit. By the tone decoding method and the device, the embodiment of the invention greatly solves the problem of uncomfortable hearing caused by the tone signals of different bandwidths and improves the hearing quality of the tone signals.

The application discloses a quantification noise reducing method and device. After an original digital signal is subjected to framing operation, digital signal sampling points with a maximum amplitude are found in two continuous frames of signals which are respectively a current frame and a next frame, and maximum object digital gain which prevents the maximum amplitude from overflowing is obtained via calculation; object sampling points having minimum own energy and adjacent energy in the current frame are used as gain switching points; according to a gain switching position and a gain switching step number, original digital gain is gradually switched to object digital gain at the gain switching points, and analog gain is correspondingly adjusted. According to the quantification noise reducing method and device, a plurality of object sampling points with the maximum energy are used as the gain switching points; thus, compared with technologies of the prior art, the quantification noise reducing method and device effectively optimize selection of the gain switching points, the gain switching points are enabled to have the minimum adjacent energy, switching noise is lowered to the minimum, and therefore quantification noise is lowered to the minimum and audio quality of signals can be improved.

The invention provides a bone conduction speech enhancement method based on joint dictionary learning and sparse representation. In a training stage, in an indoor noise-free environment, a special-shaped double-microphone array composed of a bone conduction microphone and an air conduction microphone is used for synchronously collecting training speech, and a joint training set of the bone conduction speech and the air conduction speech is constructed; and short-time inverse Fourier transform is performed on the training signals of the bone conduction speech and the air conduction speech to obtain a time-frequency spectrum amplitude, and a joint speech dictionary of the bone conduction speech and the air conduction speech is learnt on a time-frequency spectrum. In a detection stage, short-time Fourier transform is performed on the bone conduction speech to obtain a time-frequency spectrum amplitude and a phase; the amplitude is projected on a bone conduction speech sub-dictionary of the joint speech dictionary; the air-guided speech time-frequency spectrum amplitude is reconstructed by using the obtained sparse representation coefficient and the air-guided speech sub-dictionary ofthe joint speech dictionary, two methods are provided for enhancing the time-frequency spectrum of the bone conduction speech, and finally short-time inverse Fourier transform is performed to obtain an enhanced bone conduction speech time-domain signal, so that the speech sharpness is improved.

A hearing aid and a method performed by a hearing aid including at least one microphone unit, an accelerometer unit, and a processor. The method performed by the hearing aid includes generating a processed microphone signal including attenuating the microphone signal in accordance with an attenuation value; wherein the attenuation value is based on a measure of correlation between a microphone signal from the microphone unit and an acceleration signal from the accelerometer unit. The hearing aid instantly reduces noise arising from mechanical handling of the hearing aid based on input from the microphone unit and the accelerometer unit.

The invention belongs to the technical field of artificial intelligence and neural networks, and relates to a convolutional recurrent neural network voice noise reduction method based on a self-attention mechanism, which comprises a training stage and an enhancement stage. In the training stage, firstly, noise and pure voice are superposed to synthesize noisy voice, then feature extraction is performed on the noisy voice, then the pure voice and the noisy voice after feature extraction are sent into a convolutional recurrent neural network model to learn a mapping relation between the noisy voice and the pure voice, and after the training stage is finished, the pure voice and the noisy voice are subjected to feature extraction to learn the mapping relation between the noisy voice and the pure voice. Storing the trained model; and the enhancement stage comprises the steps of firstly performing feature extraction on noisy voice, then sending the noisy voice into the trained model for voice noise reduction, and finally outputting the voice after noise reduction. And effective separation of voice and noise is realized. And the separated signal is restored to a clear voice dimension, so that the whole voice noise reduction process is completed.

The invention relates to a method for automatically amplifying the volume, comprising: A, according to the parameter of distortion rate, calculating the maximum amplitude value which meets the preset distortion rate, then, dividing the maximum of sampling point with said maximum amplitude value to attain the maximum magnification times; B, individually multiplying each sampling point of said audio processing block with the maximum magnification times attained in the step A to attain the new amplitude value to be output. The inventive method via setting the parameter of distortion and neglecting the sampling point with super-high amplitude value calculate the maximum magnification times, utilizing which to process the volume amplifying, therefore it can confirm the amplifying effect of most audio flow and improve the audio amplifying quality; in addition, via the parameter of maximum magnification times, making calculated maximum magnification times less than said parameter of maximum magnification times, for avoiding the noise in the mute stage of amplifying process and improve the hearing quality in said process.

Provided in the embodiments of the present invention are a method and a device for signal encoding and decoding. The method comprises: determining, on the basis of the number of available bits and of a first saturation threshold value i, a number k of sub-bands to be encoded, where i is a positive number and k is a positive integer; selecting k sub-bands from among all the sub-bands on the basis of the quantized envelope of each such sub-band, or on the basis of a psychoacoustic model; performing a one-pass encoding operation on the spectral coefficients of the k sub-bands. In the embodiments of the present invention, a number k of sub-bands to be encoded is determined on the basis of the number of available bits and of a first saturation threshold value, and k sub-bands are selected for encoding from among the several sub-bands rather than encoding the entire frequency band; this reduces the number of spectrum holes in signals to be decoded, thereby increasing the audio quality of output signals.

The invention relates to a method for eliminating indoor reverberations, and belongs to the technical field of signal processing. The method relates to a later period reverberation power spectrum estimation module, a spectrum subtraction module, a voice / voice-free detection module, an energy decrement module and a self-adaptation spectrum line enhancement module. Input of the later period reverberation power spectrum estimation module is the reverberation voice, output of the later period reverberation power spectrum estimation module is connected with the spectrum subtraction module, input of the spectrum subtraction module is the reverberation voice and the output of the later period reverberation power spectrum estimation module, output of the spectrum subtraction module is connected with the voice / voice-free detection module, output of the voice / voice-free detection module controls the output of the spectrum subtraction module, and the output of the voice / voice-free detection module controls the output of the spectrum subtraction module to be selectively connected with the energy decrement module or the self-adaptation spectrum line enhancement module. The energy decrement module or the self-adaptation spectrum line enhancement module outputs the final enhanced voice.

A predicting method and apparatus for frequency domain pulse decoding and a decoder are provided. The method includes: dividing current frame and previous frame into spectral blocks according to spectral coefficient of the previous frame (11), judging whether the prediction between frames for the spectral block divided in the current frame is needed according to the correlation between the spectral blocks divided in the current frame and previous frame (12), for the spectral block of the current frame which is judged that the prediction between frames is needed, predicting the spectral coefficient which is not decoded of the spectral block of the current frame by using the decoded spectral coefficient in the spectral block corresponding to the previous frame and the decoded spectral coefficient in the current frame (13)..

The embodiment of the present invention discloses a method and device for recovering noise components of speech and audio signals. The method includes: receiving a code stream, decoding the code stream to obtain a speech and audio signal; determining the first speech and audio signal according to the speech and audio signal; determining The sign of each sampling value in the first speech and audio signal and the amplitude value of each sampling value; determine the adaptive normalization length; according to the adaptive normalization length and each of the sampling values The amplitude value determines the adjustment amplitude value of each sampling value; and determines the second speech and audio signal according to the sign of each sampling value and the adjustment amplitude value of each sampling value. In the embodiment of the present invention, for the voice and audio signal with rising edge or falling edge, when the noise component is restored, the signal after the noise component of the voice and audio signal is restored will not have an echo, and the auditory quality of the signal after the noise component of the voice and audio signal is restored is improved. .

Embodiments of the present invention provide signal encoding and decoding methods and equipment. The method includes: determining the number k of subbands to be encoded according to the number of available bits and the first saturation threshold i, wherein i is a positive number and k is a positive integer; selecting k from each subband according to the quantized envelope of each subband sub-bands, or select k sub-bands from each sub-band according to the psychoacoustic model; perform an encoding operation on the spectral coefficients of the k sub-bands. In the embodiment of the present invention, by determining the number k of subbands to be encoded according to the number of available bits and the first saturation threshold, and selecting k subbands from each subband for encoding instead of encoding the entire frequency band, the decoding can be reduced. The spectral hole of the signal can improve the auditory quality of the output signal.

The invention discloses a noisy speech signal processing method, device and server, belonging to the technical field of communication. The method includes: according to the silent section of the noisy speech signal, obtaining a noise signal in the noisy speech signal; for each frame in the speech signal, obtaining each frame of the speech signal according to the noise signal and the noisy speech signal The power spectrum iteration factor of a frame; according to the power spectrum iteration factor of the noisy speech signal, each frame of the noise signal and the previous frame, calculate the intermediate power spectrum of each frame of the speech signal; according to each frame of the speech signal The intermediate power spectrum and the noise signal, calculate the signal-to-noise ratio of each frame in the noisy speech signal; according to the signal-to-noise ratio of each frame in the noisy speech signal, each of the noisy speech signal and the noise signal frame to obtain the processed noisy speech signal in the time domain. The invention processes the noisy speech signal through the power spectrum iteration factor, thereby improving the auditory quality of the user.

The invention provides a speech enhancement system based on a time modeling generative adversarial network, which belongs to the technical field of speech signal processing and comprises a data acquisition unit used for acquiring noisy speech signals and downsampling the noisy speech signals; and the signal enhancement unit is used for inputting the noisy voice signal into a generative adversarial network based on time modeling, compressing and extracting a global time domain feature of the voice signal, linking the time domain feature and random noise into a feature vector, and decoding the feature vector to obtain an enhanced voice signal. According to the method, the problem that time dependence and global consideration of voice time domain features are insufficient is solved, the noise influence in voice signals is reduced, and therefore the auditory quality of enhanced voice is improved.

The invention discloses a bridge operation method for a digital telephone conference, which is characterized in that: a time delay and jitter processing mechanism is used to shape the code flow of multiple network segments; a VAD voice activation detection and a combination of single frame detection and long time window detection are adopted The sampling rate matching algorithm reduces the number of invalid channels entering the bridge operation and reduces the misjudgment rate of the operation; the short-term amplitude operation and funnel mixing operation are used to reduce the amount of calculation. The beneficial technical effects of the present invention are: 1) adapting to the large time delay jitter in the packet network environment, and providing code stream shaping of multiple network segments, providing continuous and uniform voice code streams for terminals, and improving the audio quality after decoding. 2) The VAD voice activation detection and sampling rate matching algorithm is adopted to reduce the number of invalid channels entering the bridge operation and reduce the misjudgment rate of the bridge operation. 3) The short-term amplitude calculation and funnel mixing method are adopted to greatly reduce the amount of calculation, avoid bridge misjudgment caused by impact interference, and improve the quality of bridge mixing.

The invention provides a speech enhancement system based on a time modeling generative adversarial network, belonging to the technical field of speech signal processing, comprising: a data acquisition unit for acquiring a noisy speech signal and down-sampling the noisy speech signal; The signal enhancement unit is used for inputting the noisy speech signal into a generative adversarial network based on time modeling, compressing and extracting the global time domain feature of the speech signal, linking the time domain feature and random noise into a feature vector, and correcting the The feature vector is decoded to obtain an enhanced speech signal. The invention solves the problems of insufficient time dependence and global consideration of speech time domain features, reduces the influence of noise in the speech signal, and improves the hearing quality of the enhanced speech.