66results about "Speech analysis" patented technology

The invention discloses a classroom behavior monitoring system and method based on face and voice recognition. The method comprises the following steps: a camera acquires the video information of students and teachers in classrooms; voice recording equipment acquires the voice information of the students and teachers in the classrooms; a main control processor preprocesses the received video information of the students and teachers and extracts the facial expression features and behavior features of the students and teachers; the main control processor processes the received voice information of the students and extracts the voice features of the students; and the main control processor processes the received voice information of the teachers, extracts the voice features of the teachers, calculates the scores of the teaching effect of the teachers, evaluates the teaching effect of the teachers according to the scores, and provides guidance suggestions. According to the classroom behavior monitoring system and method disclosed by the invention, the classroom behaviors of the teachers and students in the classrooms are observed, and thus the accuracy and objectivity of the evaluation can be increased, the teaching methods can be improved, and the teaching quality can be increased.

A method for providing visual sequences using one or more images comprising:

• receiving one or more person images of showing at least one face,
• receiving a message to be enacted by the person, wherein the message comprises at least a text or a emotional and movement command,
• processing the message to extract or receive an audio data related to voice of the person, and a facial movement data related to expression to be carried on face of the person,
• processing the image/s, the audio data, and the facial movement data, and generating an animation of the person enacting the message.

Wherein emotional and movement command is a GUI or multimedia based instruction to invoke the generation of facial expression/s and or body part's movement.

The invention discloses a payment authentication method, a device thereof and a system thereof, belonging to the computer technical field. The method comprises the following steps of receiving a payment authentication request sent by a terminal, detecting whether the identification information in the payment authentication request is same with the prestored identification information, extracting a current voice characteristic if the identification information in the payment authentication request is same with the prestored identification information, matching the current voice characteristic with a prestored voiceprint model, and sending authentication reply information for allowing a payment operation to the terminal if the current voice characteristic is successfully matched with the prestored voiceprint model. According to the payment authentication method, the device thereof and the system thereof, a current voice signal is confirmed through the voiceprint model, after the confirmation is successful, the subsequent payment operation is allowed, a problem that a server needs to send an authentication message and the operation cost is increased in the payment operation process in the prior art is solved, an effect that the payment security can be greatly raised by only using the voiceprint identification of the voice signal is achieved, and the operation cost brought by message authentication is greatly reduced.

Linear predictive speech coding system with classification of frames and a hybrid coder using both waveform coding and parametric coding for different classes of frames. Phase alignment for a parametric coder aligns synthesized speech frames with adjacent waveform coder synthesized frames. Zero phase alignment of speech prior to waveform coding aligns synthesized speech frames of a waveform coder with frames synthesized with a parametric coder. Inter-frame interpolation of LP coefficients suppresses artifacts in resultant synthesized speech frames.

A system and a method for the scalable coding of a multi-channel audio signal comprising a principal component analysis (PCA) transformation of at least two channels (L, R) of the audio signal into a principal component (CP) and at least one residual sub-component (r) by rotation defined by a transformation parameter (θ), comprising the following steps: formation of a frequency subband-based residual structure (Sf_r) on the basis of the at least one residual sub-component (r), and definition of a coded audio signal (SC) comprising the principal component (CP), at least one residual structure (Sf_r) of a frequency subband and the transformation parameter (θ).

To perform volume control matching the actual auditory sensation level of a human so as to improve the control accuracy. As shown in FIG. 1, an audio gain adjustment device 10 includes: an audio signal input section 12 that acquires an audio signal to be adjusted; a long-term gain reflection section 14 that reflects the adjustment of long-term gain in the acquired audio signal; a frequency division section 20 that divides the audio signal from the long-term gain reflection section 14 into three frequency bands; a short-term gain reflection section 30 that reflects the adjustment of short-term gain in each frequency band; a synthesis section 40 that synthesizes audio signals output from the short-term gain reflection section 30; and an audio signal output section 42 that outputs the synthesized audio signal.

A method for communication and for displaying an interactive avatar or hologram corresponding to a remote party.

This invention relates to retrieval for multimedia content, and provides a program endpoint time detection apparatus for detecting an endpoint time of a program by performing processing on audio signals of said program, comprising an audio classification unit for classifying said audio signals into a speech signal portion and a non-speech signal portion; a keyword retrieval unit for retrieving, as a candidate endpoint keyword, an endpoint keyword indicating start or end of the program from said speech signal portion; a content analysis unit for performing content analysis on context of the candidate endpoint keyword retrieved by the keyword retrieval unit to determine whether the candidate endpoint keyword is a valid endpoint keyword; and a program endpoint time determination unit for performing statistics analysis based on the retrieval result of said keyword retrieval unit and the determination result of said content analysis unit, and determining the endpoint time of the program. In addition, this invention also provides a program information retrieval system. With present invention, program information regarding a program attended by user can be rapidly obtained.

An apparatus for processing an audio signal and method thereof are disclosed. The method comprises receiving a downmix signal, object information indicating attribute of the object and including object number information, preset information to render the downmix signal, external preset information being inputted from external, and applied object number information indicating the number of object being applied the external preset information; determining whether the applied object number information is identical to the object number information; and rendering the downmix signal by using the external preset information, if the applied object number information is identical to the object number information, wherein the external preset rendering parameter renders the object being included in the downmix signal and the external preset metadata indicates attribute of the external preset rendering parameter.

Accordingly, an audio signal can be efficiently reconstructed by individually selecting and applying external preset information by a data region unit or by selecting and applying the same external preset information to a whole downmix signal.

Methods and apparatus are provided for coding and decoding a digital audio signal. Decoding includes: decoding according to an inverse transform decoding of a previous frame of samples of the digital signal, which is received and coded according to a transform coding; and decoding according to a predictive decoding of a current frame of samples of the digital signal, which is received and coded according to a predictive coding. The predictive decoding of the current frame is a transition predictive decoding which does not use any adaptive dictionary arising from the previous frame. At least one state of the predictive decoding is reinitialized to a predetermined default value, and an add-overlap step combines a signal segment synthesized by predictive decoding of the current frame and a signal segment synthesized by inverse transform decoding, corresponding to a stored segment of the decoding of the previous frame.

Provided is a complex exponential modulation filter bank which can reduce quantity of arithmetic operation, and can realize low electric power consumption or speeding-up. This complex exponential modulation filter bank has a step of calculating a first intermediate signal from an input signal, a step of calculating a second intermediate signal from the first intermediate signal, a step of calculating a third intermediate signal from the second intermediate signal with fast Fourier transform, and a step of calculating a complex band output signal from the third intermediate signal.

The invention discloses a method for recording synchronization, a device for the recording synchronization and equipment for the recording synchronization and belongs to the field of terminal equipment. The method comprises the steps that an audio data stream is obtained in the process of recording; the obtained audio data stream is coded according to a preset coded format and data generated by coding are written in an audio file; the data are read from the audio file which involves in the process of writing; the data which are read each time are uploaded to a server, so that the synchronization is conducted by the server according to the received data. The device comprises an audio data stream obtaining module, a writing module, a data reading module and an uploading module. Real-time recording synchronization is achieved, the effects that storage can be conducted at the moment of recording and uploading can be conducted at the moment of the storage are achieved, recording content can be recovered to the maximum extent according to the conducted synchronization on the server when the recording is not finished and interrupted by accident or deleted by accident, safety of the audio file is protected, the stability of the audio file is improved and the purpose of recovering the data is achieved.

The scheme relates to artificial intelligence, and provides an emotion recognition method, a device and a computer readable storage medium, andthe method comprises the following steps of: receiving audio data, generating a voice spectrum graph aiming at the audio data, inputting the voice spectrum graph to a first recognition module, inputting the audio data to a second recognition module, whereinthe first recognition module extracts a first feature vector by adopting a DCNN network and an RNN network which are sequentially connected the second recognition module extracts a MFCC from audio data, andthe MFCC is transformed into a second feature vector by s non-linear transformation;and the first and second feature vectors output from the first and second recognition modules are connected to form a combined feature and be input into a full connection layer and a soft Max layer in turn for emotional recognition.. By means of the scheme, the combination of the DCNN and the RNN is beneficial for simulating subtle local emotion clues, the MFCC describes voice characteristics in a short time, more emotion related features are captured through two features extraction modes, and the emotion recognition rate is improved.

The present invention provides a hearing aid capable of detecting contact vibration noise from a collected sound signal. A hearing aid (100) is provided with two microphones (110-1, 110-2), a vibration component extracting unit (120) which extracts from collected sound signals respectively obtained by the two microphones (110-1, 110-2) an uncorrelated component between two collected sound signals as a vibration component for each frequency band, a vibration noise identifying unit (130) which determines whether or not a contact noise occurs based on the vibration component for each frequency band extracted by the vibration component extracting unit (120), an acoustic signal processing unit (140) which, when generating an acoustic signal by hearing aid processing of the two collected sound signals, processes the acoustic signal depending on the presence or absence of the occurrence of the contact vibration noise, and a receiver (150) which converts the acoustic signal to sound.

The invention relates to a fraud detection method and device for video data, computer equipment and a storage medium. The method comprises the following steps: acquiring to-be-detected video data; extracting image data of each video frame from the to-be-detected video data, and dividing the image data into a plurality of image data sets according to the time sequence of each video frame, the imagedata sets which comprises image data corresponding to continuous video frames; inputting each image data set into a pre-trained image feature extraction model to obtain an image feature vector; extracting voice data from the to-be-detected video data, and obtaining a voice feature vector of the voice data; performing cascade splicing on the image feature vector and the voice feature vector to obtain a multi-modal feature vector; and inputting the multi-modal feature vector into a pre-trained fraud detection model to obtain a fraud detection result corresponding to the to-be-detected video data output by the fraud detection model. By adopting the method, the characteristic information amount can be increased, the comprehensiveness and diversity of the characteristic information are improved, and the accuracy of video data fraud detection is effectively improved.

The invention discloses a microphone array-oriented channel attention weighted speech enhancement method. The method comprises the following steps: respectively acquiring a path of noisy speech time domain signal from each microphone in a microphone array; constructing a multi-channel speech enhancement model; taking a short-time Fourier transform spectrum of each path of noisy speech of the microphone array as an input feature of a multi-channel speech enhancement model, and pre-training the multi-channel speech enhancement model; and inputting multiple paths of noisy speech acquired by the microphone array into the multi-channel speech enhancement model, and realizing microphone array speech enhancement by using the multi-channel speech enhancement model. According to the method, a UNETnetwork is adopted to process the voice frequency spectrum features, the parameter quantity is small, the calculation complexity is low, attention mechanism weighting operation is adopted for the channel dimension on the basis, attention weighting is carried out on each layer of encoder and decoder in the channel dimension, the features are weighted in deep structure simulation beam forming, and the performance of speech enhancement is improved.

The invention discloses a method and device for reducing echoes, and communication equipment. The communication equipment is provided with at least two microphones, wherein the distance between the first microphone and a loudspeaker is greater than the distance between the second microphone and the loudspeaker. The method comprises the steps: carrying out the filtering of a first echo signal s1(t) and/or a second echo signal s2(t); obtaining a target signal d(t) corresponding to the first echo signal s1(t), and a reference signal r(t) corresponding to the second echo signal s2(t), wherein the first echo signal s1(t) is an echo signal generated when the loudspeaker plays a downlink signal x(t) picked by the first microphone, and the second echo signal s2(t) is an echo signal generated when the loudspeaker plays the downlink signal x(t) picked by the second microphone; carrying out the filtering of the reference signal r(t) through employing an adaptive filter, and obtaining a filtering signal y(t); solving the difference between the filtering signal y(t) and the target signal d(t), and obtaining and outputting a residual signal e(t). The method makes the most of the signal picked by the microphone closer to the loudspeaker. Compared with the prior art, the method is better in echo inhibition effect.

This method includes steps of: A, according to the parameter bit distributing and quantizing (BDQ) the last frame signal and also meeting the coding requirement, obtaining the initial value (IV) of the parameter for BDQ the current frame signal (CFS); B, using IV to BDQ repeatedly CFS until finding the parameter meeting the requirement of coding the current frame; BDQ the found parameter to obtain the audio specimen output. Besides, this invention also discloses a corresponding audio coding device. This method and device well uses the energy and frequency component relativity between two neighboring frame signals, therefore reduces the circular number and the calculating work.

Presented herein are techniques for dynamically setting, in real-time, a ratio of acoustical stimulation signals to electrical stimulation signals delivered by a hearing prosthesis. The ratio of the acoustical stimulation signals to the electrical stimulation signals is set based on one or more characteristics or attributes of the input sound signals that are received and processed by the hearing prosthesis in order to generate the acoustical and electrical stimulation signals.

The invention discloses an anti-tonal modification interference audio fingerprint extraction method. The method comprises the following steps of framing and windowing an audio signal, performing Fourier transform, and collecting Fourier coefficients corresponding to all frames of signals; calculating energy segment indexes corresponding to all the frames of signals by using an energy segment indexcalculation method based on a peak point in the collected Fourier coefficients; and calculating energy segment values by using the energy segment indexes and filtering energy segments by a two-dimensional filtering core to extract fingerprints. By means of the method, the defect that an existing Philips fingerprint cannot resist tonal modification interference can be overcome; and on the premiseof keeping the sub-fingerprint continuity of the original Philips fingerprint and the robustness of various noise interferences, the resistance to the tonal modification interference is improved.

A method is described that estimates noise in an audio signal. An energy value for the audio signal is estimated and converted into the logarithmic domain. A noise level for the audio signal is estimated based on the converted energy value.

The embodiment of the invention relates to feeling-based multimedia processing. The invention discloses a method used for processing multimedia data. The method comprises the following steps: on the basis of multiple clusters, automatically determining user feelings in regard to the fragments of the multimedia data, wherein the multiple clusters and pre-defined user feelings are obtained in an associated way; and processing the fragments of the multimedia data at least partially on the basis of the determined user feelings in regard to the fragments. The invention also discloses a corresponding system and a computer program product.

A signal processing device comprises: a band detecting means for detecting a frequency band which satisfies a predetermined condition from an audio signal; a reference signal generating means for generating a reference signal in accordance with a detection band by the band detecting means; a reference signal correcting means for correcting the generated reference signal on the basis of a frequency characteristic thereof; a frequency band extending means for extending the corrected reference signal up to a frequency band higher than the detection band; an interpolation signal generating means for generating an interpolation signal by weighting each frequency component within the extended frequency band in accordance with a frequency characteristic of the audio signal; and a signal synthesizing means for synthesizing the generated interpolation signal with the audio signal.