107 results about "Sound source identification" patented technology

An apparatus for a speech recognition based on source separation and identification includes: a sound source separator for separating mixed signals, which are input to two or more microphones, into sound source signals by using independent component analysis (ICA), and estimating direction information of the separated sound source signals; and a speech recognizer for calculating normalized log likelihood probabilities of the separated sound source signals. The apparatus further includes a speech signal identifier identifying a sound source corresponding to a user's speech signal by using both of the estimated direction information and the reliability information based on the normalized log likelihood probabilities.

The embodiment of the invention provides a method and electronic equipment for recognizing orientation and identification, and relates to the field of computers. The method and electronic equipment for recognizing the orientation and the identification can improve the precision of recognition on a speaker, improve the efficiency of the recognition, and achieve real-time feedback. According to the method, a first orientation collection of sound sources is obtained firstly through sound source location, a second orientation collection of the sound sources is then obtained by locating people with lip actions, an identification collection of the people with the lip actions is obtained, then a third orientation collection containing the orientation information of the speakers is obtained according to the first orientation collection and the second orientation collection, and finally the orientation and the identification of the speaker is determined according to the third orientation collection and the identification collection. The method and electronic equipment for recognizing the orientation and the identification are applied to locating of the sound sources and the recognition of the sound source identification.

The present invention relates to a sound source separation and display method and a system thereof, and provides in particular a sound source separation and display method and a system thereof that are intended to eliminate a specific sound source. In order to separate a plurality of sound sources by using a single set of microphone array, the result of processing of sound source identification is utilized. More specifically, a signal in a direction is extracted from the result of the processing of the sound source identification, and a field limited to/eliminated of the effect of the signal is calculated and displayed. Such an operation can be repeated. A virtual reference signal can be created in a time domain.

A sound direction estimation apparatus includes a sound source model storage unit that stores likelihood of a sound source class in correspondence with a sound feature quantity, a sound signal input unit that receives a sound signal, a sound source identification unit that identifies a sound source class of the sound signal input from the sound signal input unit with reference to the sound source model storage unit based on the sound feature quantity of the sound signal, and a first sound source localization unit that estimates a sound direction of the sound signal of the sound source class identified by the sound source identification unit.

A system for characterizing, synthesizing, and/or canceling out acoustic signals from inanimate and animate sound sources. Electromagnetic sensors monitor excitation sources in sound producing systems, such as animate sound sources such as the human voice, or from machines, musical instruments, and various other structures. Acoustical output from these sound producing systems is also monitored. From such information, a transfer function characterizing the sound producing system is generated. From the transfer function, acoustical output from the sound producing system may be synthesized or canceled. The systems disclosed enable accurate calculation of transfer functions relating specific excitations to specific acoustical outputs. Knowledge of such signals and functions can be used to effect various sound replication, sound source identification, and sound cancellation applications.

A sound source identification apparatus includes a sound collection unit including a plurality of microphones, a sound source localization unit configured to localize a sound source on the basis of an acoustic signal collected by the sound collection unit, a sound source separation unit configured to perform separation of the sound source on the basis of the signal localized by the sound source localization unit, and a sound source identification unit configured to perform identification of a type of sound source on the basis of a result of the separation in the sound source separation unit, and a signal input to the sound source identification unit is a signal having a magnitude equal to or greater than a first threshold value which is a predetermined value.

The invention discloses a double-measuring surface noise source identification system based on near field acoustic holography, and belongs to the technical field of vehicle noise source identification systems based on near field acoustic holography. The system includes a signal acquisition module, a data storage module, an analysis processing module and a result displaying module which are connected successively. The signal acquisition module is intended for determining array position, array element spacing, sampling frequency, and calibration of sampling number and microphones, and also for measuring sound pressure signals under all working stations. The data storage module is intended for storing the measured sound pressure signals and constructing a sound pressure signal database. The analysis processing module analyzes and processes the measured sound pressure signals by adopting the signal processing method. The result displaying module uses the double-measuring surface near field acoustic holography sound source identification algorithm to reconstruct a sound field so as to obtain a sound field distribution map and acquire sound source information. The system has strong suitability to measuring environments, can avoid the problems of tradition measuring methods, such as susceptibility to explicit influence from environment noise and low rate in identifying sound source.

Disclosed is an active-noise control device including two passive attenuation earphones each provided with an external microphone, an internal microphone and a loudspeaker. The control device includes a first processing chain having a feedforward filter, a feedback filter and a reconstruction module. The control device includes a second processing chain having a sound source identification module, said second processing chain being implemented in parallel with the first processing chain, and being suitable for parameterizing the first processing chain.

The invention discloses a point acoustic source identification method without phase information, which comprises the following steps: measuring acoustic pressure radiated from a point acoustic source s at (xs,ys) on an acoustic source plane to a microphone array plane by a microphone array to obtain an acoustic pressure amplitude vector Ps; constructing an acoustic pressure amplitude vector Ps' formed on a measurement plane by a point acoustic source s' at any position on the acoustic source plane to obtain a normalization vector e=Ps'/||Ps'||; and performing inner product computation on the vector Ps and the normalization vector e: ||||=||Ps||*||e||, ||Ps||*||e||=||Ps||, ||Ps||=||||max, wherein when the maximum value ||||max is obtained, the position of a corresponding point acoustic source s' in the normalization vector e is that of the point acoustic source s on the acoustic source plane, and the strength of the acoustic source A=||||max/||Ps'|| is obtained. Therefore, the point acoustic source can be identified precisely only by measuring the acoustic pressure amplitude value.

The invention relates to an umbrella-shaped reconfigurable three-dimensional sensor array comprising a main array body and microphones installed on the main array body detachably. The main array body includes a mains shaft arranged horizontally; fixed ends of a plurality of connecting rods are arranged on the main shaft in a slidable mode; and free ends of the connecting rods radiates around the main shaft. A support rod for supporting is arranged at each connecting rod; one end of the support rod is arranged at the connecting rod in a slidable mode and the other end is hinged to the end portion of the main shaft. The main shaft, the connecting rods, and the support rods form an umbrella skeleton shape. The microphones are distributed on the connecting rods. When a sliding sleeve slides and the mounting positions of the microphones are adjusted, the microphone positions can be adjusted, so that an optimal position can be selected to obtain an optimal sound source identification and positioning effect. Moreover, a real or false sound source can be identified and a false sound source interference can be suppressed effectively.

The invention discloses a function type delay summation method for identifying a sold sphere array three-dimensional sound source, wherein a novel FDAS sound source identification method is established. In identifying incoherent sound sources in a three-dimensional space, the function type delay summation method supplies an effective approach for quickly obtaining a clean sound source imaging graph and realizes an important meaning in accurate analysis on a sound source identification result.

A method of presenting audio comprises: identifying a first ear listener position and a second ear listener position in a mixed reality environment; identifying a first virtual sound source in the mixed reality environment; identifying a first object in the mixed reality environment; determining a first audio signal in the mixed reality environment, wherein the first audio signal originates at the first virtual sound source and intersects the first ear listener position; determining a second audio signal in the mixed reality environment, wherein the second audio signal originates at the first virtual sound source, intersects the first object, and intersects the second ear listener position; determining a third audio signal based on the second audio signal and the first object; presenting, to a first ear of a user, the first audio signal; and presenting, to a second ear of the user, the third audio signal.

The invention provides a direction-of-arrival estimation method based on a differential array. An HMM-based sound source identification module and a signal source positioning module are introduced into a traditional sound source identification and positioning system; therefore, an information source signal and an interference and noise signal are separated in a complex environment; compared with atraditional positioning algorithm, the method effectively reduces the operation complexity on the basis of guaranteeing the positioning precision, and improves the discrimination among different signal sources through employing a weighted signal subspace (WSS) algorithm under the condition that the incoming wave directions of a plurality of signal sources are very close.

The invention discloses a method for identifying a planar array deconvolution sound source. The method comprises the following steps of step 1, calculating a traditional beam forming output result; step 2, establishing an equation set between the beam forming output result and the sound source distribution; and step 3, iteratively solving the sound source distribution, in the step, on the basis ofthe generalized sparsity adaptive matching pursuit gSAMP, a source intensity distribution solving method suitable for a planar array is constructed combined with a sound source identification problem. The method for identifying the planar array deconvolution sound source provided by the invention has the following technical effects of being high in spatial resolution, capable of effectively removing sidelobes and accurately locating each sound source, the positioning precision is superior to that of an existing method OMP-DAMAS, and the priori knowledge of the sparseness of a sound source signal is not needed.

The invention discloses a fast iterative shrinkage beam-forming sound source identification method, which comprises the following steps: S1, constructing a difference function; S2, calculating the minimum result error after Fourier transformation of Aq-b, and converting the non-negative least squares problem in the difference function into a minimization formula based on Fourier transformation; and S3, iteratively calculating Q, wherein Q is the distribution vector of a sound source after Fourier transformation. Compared with the existing sound source identification method, the fast iterativeshrinkage beam-forming sound source identification method of the invention has higher calculation efficiency, better convergence, faster convergence and better comprehensive performance of sound source identification.

The invention discloses a two-dimensional multi-snapshot meshless compression wave beam forming sound source identification method. The two-dimensional multi-snapshot meshless compression wave beam forming sound source identification method comprises the following steps: step 1, acquiring a measurement sound pressure matrix P*; step 2, solving sound pressure by virtue of iteration: 1, solving by using an SDPT3 solver in a CVX tool box of MATLAB: (shown in the description); 2, determining (l+1)th iteration regularization parameters kl+1; 3, obtaining a weight matrix Wl+1 determined by the iteration at (l+1)th; ending the iteration when a relative change between (shown in the description) of continuous twice iteration is smaller than or equal to 10<-3> or the maximal iteration times is completed; step 3, estimating a sound source DOA; and step 4, estimating sound source intensity. By adopting the two-dimensional multi-snapshot meshless compression wave beam forming sound source identification method, the DOA with relatively small space can be accurately estimated, the intensity of the sound source can be quantitatively acquired, and the resolution ratio, denoising capacity and soundsource identification precision can be improved.

The invention discloses a sound source identification method and device, a server and a storage medium. The sound source identification method comprises the following steps: spectrograms of voice signals are acquired, the spectrograms are spectrograms or energy spectrograms, the number of the voice signals is at least two, and the spectrograms are in one-to-one correspondence with the voice signals; by inputting the spectrograms into an identity coding model, identity coding vectors of the spectrograms are obtained; and the voice signals corresponding to at least one same sound source are summarized according to the identity coding vectors. According to the technical scheme, the effects of reducing the calculation difficulty of recognizing and summarizing the voice signals corresponding tothe at least one same sound source according to the identity code vectors of the spectrograms corresponding to the voice signals of the user and consuming resources and occupying space are achieved.And meanwhile, the accuracy of separating and classifying the voice signals generated by the plurality of sound sources is also improved.

An acoustic processing apparatus includes a sound source localization unit configured to estimate a direction of a sound source from an acoustic signal of a plurality of channels, a sound source separation unit configured to perform separation into a sound-source-specific acoustic signal representing a component of the sound source from the acoustic signal of the plurality of channels, and a sound source identification unit configured to determine a type of sound source on the basis of the direction of the sound source estimated by the sound source localization unit using model data representing a relationship between the direction of the sound source and the type of sound source, for the sound-source-specific acoustic signal.

The invention discloses a data processing method for low-frequency ultrasonic thoracic cavity imaging, which comprises the following steps: modeling through COMSOL, neglecting a spine and a heart in a model, combining two lungs into an ellipse with a long axis length of 0.15 m and a short axis length of 0.1 m, arranging a circle of annular skeletal muscle outside the ellipse of the lungs, uniformly placing 12 ultrasonic transducers around the exterior of the skeletal muscle, the material is set to be PZT-5H; one transducer is used as a sound source Sk which needs to be identified, seven transducers which directly face each other are used as receiving array elements, and signals received by the array elements are subjected to sound source identification processing by using a multi-signal classification (Multiple Signal Characterization,MUSIC) algorithm; a model is established through COMSOL, and a data processing technology applied to the field of low-frequency ultrasonic thoracic cavity imaging is creatively discovered according to a cyclic transmitting-receiving mode of one transmitting and seven receiving. According to the technology, the signal-to-noise ratio of array receiving data can be improved, and the mean square error of the data is reduced.

The invention discloses a camera shooting angle adjusting method and display equipment. A camera can rotate in a preset angle range. A controller is configured to acquire person sound source information collected by a sound collector and perform sound source identification, determine sound source angle information used for identifying the orientation angle of the position where a person is located, determine a target rotation direction and a target rotation angle of the camera based on the current shooting angle of the camera and the sound source angle information, and adjust the shooting angle of the camera according to the target rotation direction and the target rotation angle, so that the shooting area of the camera directly faces the position where the person is located. Visibly, according to the method and the display equipment provided by the invention, the rotation of the camera can be triggered by utilizing the sound source information of a person, the real-time position of the user can be automatically identified, and the shooting angle of the camera can be adjusted, so that the camera can always shoot images containing portraits.

The invention relates to the field of air conditioners, in particular to an air conditioner pipeline low-frequency noise source recognition method. The problem that the position of a low-frequency noise source of a pipeline cannot be accurately recognized through single structural field simulation can be solved, and meanwhile the accuracy of low-frequency noise source recognition is greatly improved. The method comprises the steps of calculating compressor rotational inertia moment, converting the compression rotation inertia moment; obtaining the amplitude of the rotational inertia moment ateach frequency point; generating a fluid-solid coupling finite element model of the compressor pipeline system; and loading the amplitude of the rotational inertia moment at each frequency point to afluid-solid coupling finite element model of the compressor pipeline system, calculating to obtain a frequency response curve of each section of pipeline in the compressor pipeline system, and comparing the amplitude of the frequency response curve of each section of pipeline to determine the position of the low-frequency noise source.