3474 results about "Spectrogram" patented technology

Synchronizes a secondary audio track to a video. Analyzes at least one track of a video using audio frequency analysis or spectrograms, image analysis or text analysis to find distinct audio / image / caption events from which to ensure synchronization of a secondary audio track. For example, commentary that mocks a character may be played immediately after a particular noise in the audio track of a video occurs such as a door slam. Keeping the secondary audio track in synch with the audio track of a video is performed by periodically searching for distinct events in a track of a video and adjusting the timing of the secondary audio track. May utilize a sound card on a computer to both analyze a DVD sound track and play and adjust timing of the secondary audio track to maintain synchronization. Secondary audio tracks may be purchased and / or downloaded and utilized to add humorous external commentary to a DVD for example.

A method and system for blind estimation of time and frequency parameters of an incoming signal. A spectrogram of the signal is processed using singular value decomposition (SVD) and ICA (independent component analysis) techniques. Time and frequency parameters are estimated from time and frequency eigenvectors, respectively.

Techniques for determining a feature in an image or soundtrack of one or more dimensions include receiving a subject image. A sparse transformed subject image is determined, which represents the subject image with a few significant coefficients compared to a number of values in the subject image. Multiple patch functions are received, which are based on a portion of a sparse transformed image for each of a training set of images and which represent learned features in the training set. A feature is determined to be in the subject image based on the transformed subject image and the plurality of patch functions. In various embodiments, a wavelet transformation or audio spectrogram is performed to produce the sparse transformed images. In some embodiments, the feature in the subject is determined regardless of feature location or size or orientation in the subject image.

The invention relates to an expressway audio vehicle detection device and a method thereof. By the aid of the detection device, a microphone array signal acquisition module acquires an audio signal on a lane, the audio signal is subjected to band-splitting filtering and framing through a signal processing module after being subjected to de-noising processing through a noise suppression module, cross-correlation processing is conducted among sub-band signals, an audio space spectrogram is obtained, a vehicle detection module tracks a track of the maximum value on the audio space spectrogram and judges whether a vehicle passes, and the vehicle type and the vehicle speed are obtained through a vehicle type recognition module and a vehicle speed recognition module if the vehicle passes. The detection method is based on the device, a minimum statistical noise estimation method of the adaptive window length is adopted, the signal which is subjected to noise suppression processing is subjected to band-splitting filtering and framing processing, then cross-correlation processing is conducted among same sub-band signals, cross-correlation results are summed after being subjected to amplitude compression and are unfolded along a timer shaft, and an audio signal space-time spectrum is obtained. The method and the device have the advantages of being low in cost, low in energy consumption,easy to construct, interference resisting, capable of working in all weather and the like.

This invention discloses a rare earth Y molecular sieve and preparation; wherein, the rare earth content is 12~22 w.t% considering the quantity of rare earth oxide in molecular sieve, all rare earth ion is in molecular sieve small cage; there is no peak at 0ppm in 27Al M AS NMR spectrogram. The step to obtain molecular sieve comprises: exchanging NaY molecular sieve liquid with ammonium salt or not, taking ion exchange with rare earth chloride by weight ratio of NaY dry basis:RECl3=1:0.17~0.35 at temperature 5~100Deg; pH=2.5~7.5, water: NaY=3~50; adjusting pH value to 8~11 with alkaline solution; mixing, filtering, cleaning, drying; baking for more than 0.1h with 0~100% vapor at temperature 200~950Deg; then, dealing with the molecular sieve according to weight ratio molecular sieve dry basis: ammonium salt :water=1:0~1:2~50 at temperature 60~100Deg; finally, cleaning, filtering, drying and obtaining the product.

This invention relates to an MALDI-TOF MS (Matrix-assisted Laser Desorption / Ionization Time of Flight Mass Spectrometry)-assisted identification method for listeria monocytogenes, comprising the following steps of: (1) selecting 35 to 40 strains of fresh cultures of listeria monocytogenes, pre-treating the samples; (2) collecting maps of all bacterial strain samples; (3) analyzing and unifying the obtained maps by using BioTyper software to obtain a standard map of listeria monocytogenes; (4) preparing a sample of a microorganism to be detected by adopting the method in the step (1); (5) collecting the MALDI-TOF-MS map of the sample to be detected according to the method in the step (2); comparing the obtained MALDI-TOF-MS map of the sample to be detected in step (5) with the obtained standard map of listeria monocytogenes in the step (3), judging the detection results according to matching fractions. According to the invention, a mass spectrogram database and the standard map in listeria monocytogenes identification are created successfully, and the accurate identification with convenience, microscale and high automation for listeria monocytogenes can be realized.

The invention discloses a rolling bearing sound signal fault diagnosis method based on short-time Fourier transform and a sparse laminated automatic encoder. According to the method of the invention, firstly a smart mobile phone is used for acquiring a sound signal of the rolling bearing fault; then short-time Fourier analysis is performed on the sound signal for obtaining a spectrogram matrix; then the modulus value of the matrix is acquired and gray scale normalization processing is performed; then the normalized data are selected and input into a deep studying network for automatically extracting characteristics; and finally the characteristic which is extracted by a neural net is input into a Softmax classifier for identifying the fault mode. The invention provides the rolling bearing sound signal fault diagnosis method based on smart mobile phone sound signal short-time Fourier transform (STFT) and the sparse laminated automatic encoder (SAE). Through testing result analysis, the rolling bearing sound signal fault diagnosis method can accurately determine the fault mode of the rolling bearing.

The invention discloses an estimation method of a co-prime array DOA (Direction Of Arrival) angle based on sparse reconstruction, and mainly solves the problems that a prior art is higher in operand, less in identification information source amount and large in passive location evaluated error, and needs more priori knowledge. The method comprises the realizing steps of forming a co-prime array by an antenna receiver; obtaining observation data to spatial signal sampling; receiving data vectors by a virtual array element obtained by observation data; dividing spatial grids to form over-complete bases; receiving a spare relationship between the data vectors and the over-complete bases by the virtual array element to build a spare restraint equation; resolving the spare restraint equation by adopting a convex optimization method to obtain sparest resolution; drawing a magnitude spectrogram by a relative relationship between the sparest resolution and the spatial angle to obtain a DOA angle value. According to the method provided by the invention, the passive direction-finding precision and operating speed can be improved under a condition of low priori knowledge, the number of the recognized information source can be improved, and the estimation precision of a signal direction angle can be improved in a low signal to noise ratio, therefore the estimation method can be used for target reconnaissance and passive location.

We describe a method of blind source separation for use, for example, in a listening or hearing aid. The method processes input data from multiple microphones each receiving a mixed signal from multiple audio sources, performing independent component analysis (ICA) on the data in the time-frequency domain based on an estimation of a spectrogram of each acoustic source. The spectrograms of the sources are determined from non-negative matrix factorization (NMF) models of each source, the NMF model representing time-frequency variations in the output of an acoustic source in the time-frequency domain. The NMF and ICA models are jointly optimized, thus automatically resolving an inter-frequency permutation ambiguity.

System, method, and apparatus for determining the composition of a sample of material. In one embodiment, the method pertains to the counting of photons that were inelastically scattered by the sample, and for minimizing the effects of fluorescent or phosphorescent photons. In yet another embodiment of the invention, a sample is illuminated by a repetitive pulse of monochromatic light, and the resultant scattered photons from the samples are collected and counted during a predetermined integration period. Yet other embodiments pertain to a low-cost, computer-controlled system for repetitively counting inelastically scattered photons so as to create a Raman histogram and a Raman spectrogram of the photons.

The finding of the present invention is that the melody extraction or automatic transcription may be implemented clearly more stable and if applicable even less expensive when the assumption is considered sufficiently that the main melody is the portion of a piece of music which man perceives the loudest and the most precise. Regarding this, according to the present invention the time / spectral representation or the spectrogram of an interesting audio signal is scaled using the curves of equal volume reflecting human volume perception in order to determine the melody of the audio signal on the basis of the resulting perception-related time / spectral representation.

Spectrogram extraction from DNA sequence has been known since 2001. A DNA spectrogram is generated by applying Fourier transform to convert a symbolic DNA sequence consisting of letters A, T, C, G into a visual representation that highlights periodicities of co-occurrence of DNA patterns. Given a DNA sequence or whole genomes, with this method it is easy to generate a large number of spectrogram images. However, the difficult part is to elucidate where are the repetitive patterns and to associate a biological and clinical meaning to them. The present disclosure provides systems and methods that facilitate the location and / or identification of repetitive DNA patterns, such as CpG islands, Alu repeats, tandem repeats and various types of satellite repeats. These repetitive elements can be found within a chromosome, within a genome or across genomes of various species. The disclosed systems and methods apply image processing operators to find prominent features in the vertical and horizontal direction of the DNA spectrograms. Systems and methods for fast, full scale analysis of the derived images using supervised machine learning methods are also disclosed. The disclosed systems and methods for detecting and / or classifying repetitive DNA patterns include: (a) comparative histogram method, (b) feature selection and classification using support vector machines and genetic algorithms, and (c) generation of spectrovideo from a plurality of spectral images.

The invention discloses a speech emotion recognition method. The method includes the steps that firstly, a speech signal is converted into a spectrogram to serve as initial input; secondly, a deep convolutional neural network is trained to automatically extract emotion features; thirdly, a stack type auto-encoder is trained for each kind of emotions, and all the stack type auto-encoders are fused to automatically construct membership functions of an emotion fuzzy set; fourthly, the features obtained in the second step are subjected to feature optimization by means of the fuzzy optimal theory in the third step; fifthly, emotion classification recognition is conducted by means of a Softmax classifier. The method takes abstract fuzzy properties of speech emotion information into consideration, the extracted emotion features are subjected to selective fuzzy optimization to improve the significance of the features, fuzzy membership functions in the fuzzy theory are automatically constructed by means of the concept of deep neural network layer-by-layer training, and the problem that the proper membership functions in the fuzzy theory are difficult to select and determine is solved.

The present invention discloses a method for on-line monitoring local discharge of power cable and its device. On the earthing wire of power cable a local discharge sensor is cover-connected, and a resistance divider is mounted on the voltage transformer of transformer station, and a computer can be used for making analysis and processing local discharge, and can be used for making statistial analysis. Said method can analyze maximum discharge quantity, average discharge quantity, discharge frequency, discharge phase, sewness, steepness and other discharge parameters, and can have the rich spectrogram display function, can display discharge quantity (including maximum discharge quantity and average discharge quantity) and discharge phase spectrogram, discharge frequency and discharge phase spectrogram and others. Said invention can utilize ORACLE to implement data management.

Systems and techniques are provided for content filtering with convolutional neural networks. A spectrogram generated from audio data may be received. A convolution may be applied to the spectrogram to generate a feature map. Values for a hidden layer of a neural network may be determined based on the feature map. A label for the audio data may be determined based on the determined values for the hidden layer of the neural network. The hidden layer may include a vector including the values for the hidden layer. The vector may be stored as a vector representation of the audio data.

The invention discloses a speech emotion recognition method based on parameter migration and a spectrogram, and the method comprises the following steps: 1), collecting the speech emotion data from aChinese emotion database of Institute of Automation, Chinese Academy of Sciences, and carrying out the preprocessing of the speech emotion data, wherein the speech emotion data contains six emotions:anger, fear, happiness, neutral emotion, sadness, and surprise; 2), constructing a network model based on a pre-trained convolutional cyclic neural network; 3), carrying out the parameter migration and training of the network model at step 2). The method can achieve the extraction of the emotion features of the spectrogram in the time and frequency domains, improves the recognition accuracy, alsocan learn the pre-training technology, and improves the network training speed.

The invention discloses a speaker recognition method based on a convolution neural network and a spectrogram, and the method comprises the following steps: firstly collecting an audio signal of each speaker; secondly converting the audio signals into the spectrogram; thirdly taking an image as an input layer, and training the neural network through AlexNet training; fourthly adjusting the weight values and biases of all layers of the neural network layer by layer through a reverse propagation algorithm; finally obtaining the parameters of the neural network, and classifying the speakers. The method achieves the quick recognition of the speakers through a convolution neural network processing method.

The invention discloses a speech classification method based on a deep neural network, and aims at solving different speech classification problems through a unified algorithm model. The method of theinvention includes the steps of S1, converting a speech into a corresponding spectrogram; segmenting the complete spectrogram along the frequency domain into blocks to obtain a local frequency domaininformation set; S2, taking the complete and local frequency domain information as inputs of a model respectively, and based on the different inputs, the convolutional neural network being capable ofextracting local and global features; S3, using an attention mechanism to fuse global and local feature expressions to form a final feature expression; S4, using tagged data to train the network by gradient descent and back propagation algorithms; and S5, using the trained parameters for an untagged speech and taking the classification of highest probability that the model outputs as a predictionresult. The method of the invention realizes a unified algorithm model for different speech classification problems, and improves the accuracy on multiple speech classification problems.

The invention discloses an infrared cavity ring-down spectroscopy trace gas detection method based on quantum cascade laser; comprising the following steps: using a tunable quantum cascade laser as alight source, and selecting measuring waveband and wavelength scan step length aiming at spectrum line characteristic of the gas to be detected; respectively measuring the cavity ring-down time of each wavelength in the cavity with absorption or without absorption by cavity ring-down technique, and calculating the gas absorption coefficient of corresponding wavelength so as to obtain a relation curve that is an absorption spectrogram of the measured gas absorption coefficient and toned laser wavelength. The spectrogram is contrasted with spectrum line characteristic of corresponding gas in HITRAN database, thereby being capable of analyzing and determining weather the measured gas contains the predetermined gas component; the absolute concentration of the gas to be measured can be calculated and obtained and the absolute concentration of the gas to be measured can be measured through scaling measurement by using absorption peak wavelength of the absorption spectrum as the best detection wavelength and the relation among the gas absorption coefficient of the wavelength, the absorption cross section and concentration. The method has high measuring sensitivity and high property of resisting interference so that the fast and exact on-line analysis and detection of multiple trace gases are easily realized.

Described herein are embodiments of an end-to-end text-to-speech (TTS) system with parallel wave generation. In one or more embodiments, a Gaussian inverse autoregressive flow is distilled from an autoregressive WaveNet by minimizing a novel regularized Kullback-Leibler (KL) divergence between their highly-peaked output distributions. Embodiments of the methodology computes the KL divergence in a closed-form, which simplifies the training process and provides very efficient distillation. Embodiments of a novel text-to-wave neural architecture for speech synthesis are also described, which are fully convolutional and enable fast end-to-end training from scratch. These embodiments significantly outperform the previous pipeline that connects a text-to-spectrogram model to a separately trained WaveNet. Also, a parallel waveform synthesizer embodiment conditioned on the hidden representation in an embodiment of this end-to-end model were successfully distilled.

Methods of using individually distinctive patterns of voice characteristics to identify a speaker include computing the reassigned spectrogram of each of at least two voice samples, pruning each reassigned spectrogram to remove noise and other computational artifacts, and comparing (either visually or with the aid of a processor) the strongest points to determine whether the voice samples belong to the same speaker.

The invention relates to a device and a method for detecting pipeline closed cracks based on the vibro-acoustic modulation technology and belongs to the pipeline nondestructive testing field. The device provided by the invention comprises a computer, a first function generator, a second function generator, a first power amplifier, a second power amplifier, a vibration exciter, an accelerometer, acharge amplifier, a first thickness concertina-type piezoelectric ceramic sheet array and a second thickness concertina-type piezoelectric ceramic sheet array, an anti-aliasing filter and a data collection card. The method provided by the invention comprises the following steps of: acquiring a pipeline first-order bending vibration inherent frequency by swept frequency excitation; taking the frequency as the one of low-frequency vibration in the vibro-acoustic modulation detection; simultaneously exciting low-frequency bending vibration and high-frequency supersonic waves into the pipeline, determining whether there exists closed cracks inside the pipeline by the existence of frequency components which equal high frequency supersonic wave frequency subtracts or adds low frequency bending vibration frequency in a received signal frequency spectrogram. The device provided by the invention is utilized to solve the problem that it is hard to detect closed cracks by traditional methods, and can be applied to detect closed cracks in industrial pipelines.

A selective intrusion detection system includes a Doppler transceiver configured and adapted to receive Doppler return signals indicative of moving targets present in a surveillance space. A processor is operatively connected to the Doppler transceiver to convert Doppler return signals into spectrograms and to determine whether any given spectrogram is indicative of presence of a human or another moving target, like a domestic pet. An alarm is operatively connected to the processor, wherein the processor and alarm are configured to provide an alert in the event the processor determines any given spectrogram is indicative of a human, and to forego providing an alert in the event the processor determines any given spectrogram is indicative of another moving target only.

A music-synchronized display apparatus includes a beat extractor configured to detect a portion in which a power spectrum in a spectrogram of an input music signal greatly changes and to output a detection output signal that is synchronized in time to the changing portion in synchronization with the input music signal; a tempo value estimation section configured to detect the self-correlation of the detection output signal from the beat extractor and to estimate a tempo value of the input music signal; a variable frequency oscillator in which an oscillation center frequency is determined on the basis of the tempo value from the tempo value estimation section and the phase of the output oscillation signal is controlled on the basis of a phase control signal; a phase comparator; a beat synchronization signal generation and output section; an attribute information storage section; an attribute information obtaining section; and a display information generator.

The invention provides a simulative-target metabonomics analytic method based on combination of liquid chromatography and mass spectrum. The method includes following steps: respectively manufacturing to-be-analyzed samples into merged samples according to groups; automatically collecting secondary mass spectrums of metabolites in each merged samples through data of UHPLS / Q-TOF MS with dependence on a collecting mode; extracting a retaining time and information of parent ions and daughter ions of the metabolites by qualitative analytic software; screening out characteristic ion pair information of the metabolites according to daughter ion response intensity; summing the characteristic ion pair information of each merged samples; and scanning the obtained characteristic ion pair of the metabolites by UHPLC / QQQ MS through a dynamic multi-reaction monitoring mode in an actual sample for obtaining a corresponding spectrogram; and performing integration through quantitative analytic software to obtain the metabolites in the to-be-analyzed samples and content information thereof.

The invention discloses an early warning system for an elevator fault and an early warning method for the elevator fault. The system comprises a monitoring center, a tractor sensor arranged on a tractor, an elevator car sensor on an elevator car, and a noise sensor arranged outside the elevator car, wherein the tractor sensor, the elevator car sensor and the noise sensor acquire vibration data and noise data of an elevator under different running stages, and transmit the data to the monitoring center, the monitoring center carries out fast fourier transform on the data, to obtain a vibration spectrogram and a noise spectrogram of the elevator under the different running stages, and compares with the vibration spectrogram and the noise spectrogram of the elevator under a normal running stage. The amplitude values of the frequency ranges are compared whether to be abnormal or not, so that the early warning of the elevator fault can be realized. According to the system disclosed by the invention, the fault condition of the elevator can be effectively forecast, the running safety of the elevator can be further improved, and the normal running of the elevator can be guaranteed.

Automatic recognition of sample media content is provided. A spectrogram is generated for successive time slices of audio signal. One or more sample hash vectors are generated for a time slice by calculating ratios of magnitudes of respective frequency bins from a column for the time slice. In a primary evaluation stage an exact match of bits of the sample hash vector is performed to entries in a look-up table to identify a group of one or more reference hash vectors. In a secondary evaluation stage a degree of similarity between the sample hash vector and each of the group of reference hash vectors is performed to identify any reference hash vectors that are candidates for matching the sample media content, each reference hash vector representing a time slice of reference media content.

In order to implement the melody extraction or the automatic clearly more stable or to improve the transcription result, respectively, at the resulting segments or trajectories, respectively, of a melody line gained from a spectrogram of the audio signal a harmony mapping is performed such that a follower segment directly neighboring a reference segment in time direction is virtually shifted in the frequency direction by stages of octave, fifth and / or third in order to examine whether among the resulting lines of the octave, fifth and / or third there is one that fulfills a predetermined condition, like e.g. that the time / spectral representation along this line comprises a minimum which is larger by a certain factor than a minimum that it comprises along the reference segment line, and, if such a line exists, selects the same and actually performs the shifting of the follower segment. This way, errors in melody line determination may be corrected again. By a suitable selection of predetermined conditions for the lines of the octave, fifth and / or third, the direct neighborhood and further possible conditions, a faulty shifting may be prevented.

A pulse Doppler ultrasound system and associated methods are described for monitoring blood flow. A graphical information display includes simultaneously displayed depth-mode and spectrogram displays. The depth-mode display indicates the various positions along the ultrasound beam axis at which blood flow is detected. These positions are indicated as one or more colored regions, with the color indicating direction of blood flow and varying in intensity as a function of detected Doppler ultrasound signal amplitude or detected blood flow velocity. The depth-mode display also includes a pointer whose position may be selected by a user. The spectrogram displayed corresponds to the location identified by the pointer. Embolus detection and characterization are also provided.