83 results about "Sound classification" patented technology

The embodiment of the invention provides a method and a device for identifying the authenticity of sound in audio information, which can refine classification categories under the condition of identifying that the sound in the audio information is real sound or forged sound (such as synthetic sound), and specifically, the method and the device can be used for identifying the authenticity of the sound in the audio information. A real sound or counterfeit sound classification category is each refined into at least one hidden category within a hidden space, a single hidden category being characterized by a single prototype vector. According to the method, after the audio information to be recognized is coded to obtain the corresponding coding vector, the coding vector can be compared with each prototype vector to obtain each corresponding similarity, and then the sound authenticity of the audio information to be recognized is determined according to each similarity, namely, the audio information belongs to a real sound classification category or a forged sound classification category. Therefore, the accuracy of sound authenticity identification in the audio information can be improved.

The invention provides a sound classification target model construction method and system, and relates to the technical field of sound signal processing, and the method comprises the steps: obtaining original sound data and auxiliary modal data, carrying out the sound data conversion processing of the original sound data, and obtaining corresponding multi-resolution sound image data; processing the auxiliary modal data to obtain a corresponding auxiliary modal space-time matrix; performing local feature processing on the multi-resolution sound image data and the auxiliary modal space-time matrix to obtain local sound image features and auxiliary modal features; constructing a hierarchical feature sequence space according to the local sound image features and the auxiliary modal features, and obtaining a corresponding fusion feature sequence; and constructing a sound classification target model according to the fusion feature sequence, and generating a dynamic sound index according to the sound classification target model, thereby obtaining a sound classification result, and improving the accuracy of sound classification.

The invention provides a micromotor abnormal sound classification method and device based on multi-scale feature fusion and an attention mechanism. The micromotor abnormal sound classification method comprises the following steps: 1, acquiring sound signal data; 2, sound signal preprocessing; step 3, sound data feature extraction; 4, performing multi-scale feature fusion: performing multi-level feature information extraction on the comprehensive feature vector by adopting convolutional neural sub-networks of three different-scale convolution kernels to obtain a feature map; a channel attention mechanism and a space attention mechanism are integrated in each convolutional neural sub-network, the importance of each channel feature vector is weighted through the channel attention mechanism, and the region of interest of the feature vector is dynamically adjusted in the spatial dimension through the space attention mechanism; and 5, performing feature classification and judging the running state or the fault type of the current micromotor. According to the method, fusion of multi-scale convolution and an attention mechanism is introduced, the robustness of the model in a complex environment is enhanced, and the accuracy of fault classification is improved.

The invention discloses a small sample registered regional bird chirp classification method and related equipment, and the method comprises the steps: inputting a first to-be-classified Fbank feature into a bird voiceprint feature extraction model, generating a to-be-classified bird voiceprint template feature, and carrying out the updating of a small sample registration template library according to the to-be-classified bird voiceprint template feature; inputting the second to-be-classified Fbank feature into a bird voice voiceprint feature extraction model, and extracting a to-be-classified voiceprint embedding feature; and performing cosine similarity calculation on the to-be-classified voiceprint embedded features and the voiceprint features in the updated small sample registration template library to obtain a to-be-classified feature matching score set, and selecting a bird corresponding to the highest matching score as a regional bird buzzing classification result. The method can improve the feature extraction robustness in a field complex noise environment, improves the accuracy of regional bird recognition, does not need to train a model again, greatly reduces the extension cost, and can be widely applied to the technical field of sound signal recognition.

The invention discloses a breath sound electronic auscultation signal identification method based on multi-mode and deep learning, and relates to the technical field of deep learning and multi-mode learning. Firstly, audio data related to breathing sound classification and text description corresponding to the audio data are obtained; preprocessing the audio data, including de-noising and feature extraction; thirdly, matching an audio sample with the generated text description through a comparative learning method so as to optimize the representation capability of the model; evaluating and verifying the trained model, and performing label prediction on an audio sample by using a classification head; and finally, combining the metadata with the audio features by adopting a multi-modal fusion method, and carrying out classification prediction through a reasoning stage. On the basis of focusing on the classification accuracy of the model, the multi-modal input effect of the model is focused at the same time, the robustness of the model is improved by integrating the metadata, and whether the model can still keep good classification performance or not under the condition that part of the metadata is missing is verified. In addition, the method also reveals the importance of different metadata on the model prediction result and the interaction relationship of the different metadata. In conclusion, the invention can provide reliable support for clinical application.

The invention relates to the technical field of audio signal processing and biomedical signal analysis, and still has a further optimized space for the recognition of anti-noise requirements, signal individual differences and complex pathological modes in a noise environment. The invention provides a double-path CNN heart sound classification method based on time-frequency and double-spectrum fusion features, and the method comprises the steps: carrying out the preprocessing of an original heart sound signal of a data set which is classified into a normal heart sound and an abnormal heart sound, and obtaining a to-be-recognized heart sound signal; based on dynamic continuous wavelet transform, adaptively selecting parameters to extract time-frequency characteristics, introducing bispectrum analysis, capturing nonlinear characteristics, generating a dual-channel characteristic pattern, and efficiently storing the dual-channel characteristic pattern in an HDF5 format; and based on a designed double-path convolutional neural network structure, respectively processing the extracted time-frequency and double-spectrum features, performing classification after fusion, and training a model in combination with category weighted loss and an optimization strategy to obtain a heart sound classification result. The heart sound recognition accuracy can be improved.

An anti-snoring system comprising an adjustable bed having a sleeping surface that may be mechanically raised or lowered, and a control module adapted to receive commands from a source external to the adjustable bed; a mobile device in direct or indirect communication with the control module in the adjustable bed, and wherein the mobile device has sound recording capabilities; a mobile application resident on the mobile device, wherein the mobile application includes a sound classification machine learning model that includes an artificial intelligence or neural network operative to determine whether or not a person on the sleep surface is snoring, and wherein upon a determination that the person is snoring and has been snoring for a predetermined period of time, the mobile application instructs the control module to raise or adjust the sleeping surface to a height or position that will discourage the person from snoring.

The invention discloses an environmental sound classification method and model based on neural evolution, a medium and equipment. The method comprises the following steps: S000, searching a feature hyper-parameter, a convolutional neural network hyper-parameter and a cross-layer connection structure in a preset feasible region; s100, optimizing a hyper-parameter and a cross-layer connection structure, wherein S110, initializing an algorithm; s120, coding the cross-layer connection structure and sending the cross-layer connection structure and the hyper-parameters to the network model; s130, calculating the fitness of an optimization result; s140, obtaining an optimization result by adopting a neural evolutionary algorithm; s150, judging, if an iteration condition is met, executing S120, otherwise, updating an optimization result, and S200, constructing and training a network model according to the optimization result, including S210, acquiring a training set and a test set; s220, carrying out feature extraction; s230, processing by using a convolutional neural network; and S240, the classification result score is output to S130.

The invention relates to the technical field of artificial intelligence, and discloses an environmental sound classification method and device, computer equipment and a storage medium in the fields of financial science and technology and medical health. The environmental sound classification method comprises the following steps: acquiring real-time environmental sound data, and preprocessing the real-time environmental sound data to obtain a sound sample to be analyzed; performing data analysis processing on the to-be-analyzed sound sample through a preset classification model to obtain a sound classification result corresponding to the real-time environment sound data; the preset classification model is obtained by training an initial classification model through an initial sound sample and a synthetic sound sample, the synthetic sound sample is obtained by performing data synthesis processing on the initial sound sample through a preset synthesis model, and the preset synthesis model comprises a generative adversarial network and a diffusion model network. According to the method, the data processing efficiency and the sample quality of model training can be improved, the generalization ability and robustness of the model are improved, and the accuracy and stability of environmental sound classification are improved.

Marine mammal sound lacks training data with annotations, so that a large-scale neural network model trained in a supervised mode is difficult to use at present. Inspired by the overlapping frequency of human sound and marine mammal sound, the Man2Marine method is invented, a large amount of unannotated human sound is used for pre-training a self-supervised large model, and then the sound of the marine mammal is used for fine tuning of the self-supervised large model. Meanwhile, in order to solve the problems of excessive parameterization and difficulty in application in a pre-training self-supervision large model, a knowledge distillation technology is used for compressing model parameters. According to the method, the cost of acquiring the sound data of the marine mammals is obviously reduced, and a method with practical application value is provided for research on the sound of the marine mammals.

A method of training a sound classification model for a premises monitoring system may include receiving audio data corresponding to a sound detected at a premises. The audio data may be provided to an active instance of a sound classification model, which may generate classification data indicating the sound is unrecognized. The audio data may be provided to a user device, and updated classification data indicating an identity of the sound may be received from the user device. The audio data and the updated classification data may be stored in a data bucket corresponding to the identity of the sound. When the data bucket contains a threshold quantity of user-classified audio data, a new instance of the sound classification model may be generated and retrained using the user-classified audio data. The active instance of the classification model may be replaced with the new instance of the sound classification model.

The invention discloses a single-wheel self-balancing rail inspection device and method based on sound detection. The single-wheel self-balancing rail inspection device comprises a single-wheel self-balancing rail inspection device body and a server terminal. The single-wheel self-balancing track inspection device is placed on a rail, the posture and the balance state of the inspection device are detected through a gyroscope in the moving process, a rail image is recognized through a tracking camera, and the single-wheel self-balancing track inspection device is controlled to keep a self-balancing state so as to perform tracking movement along the single rail. In the moving process, the rail is knocked at fixed time or fixed distance, sound is collected, sound data is transmitted to the server terminal for real-time sound detection, and the working condition of the rail is distinguished according to sound classification. According to the invention, the inspection device is driven by means of camera tracking, self-balancing control and single-wheel advancing, so that the inspection device can walk on a single rail and can be kept stable; meanwhile, sound generated by knocking the rail is collected, the sound is detected through a machine learning or deep learning method, the sound is recognized in a multi-classification mode, the rail working condition is judged, and sound-based rail abnormal working condition detection is achieved.

The invention relates to a borborygmus detection system, belongs to the technical field of borborygmus detection, and solves the problem of inaccurate borborygmus recognition in the prior art. The system comprises a signal acquisition device which comprises a laser, a distributed optical fiber sensor and a collector; the laser is used for periodically emitting laser to the distributed optical fiber sensor; the collector is used for collecting a signal returned by the optical fiber sensor and sending the signal to the borborygmus resolving module; the borborygmus resolving module is used for resolving the acquired signal based on the optical frequency domain reflection, judging whether borborygmus occurs or not, and if so, determining the generation position of the borborygmus and extracting a borborygmus signal; and the borborygmus classification module is used for obtaining the borborygmus type of the individual based on a trained borborygmus classification model according to the generation position of the borborygmus and the borborygmus signal. Therefore, the borborygmus can be accurately identified.

The invention relates to an intelligent fan abnormal sound distinguishing and detecting method and system based on a time-frequency diagram and deep learning. The method comprises the steps that a fan sound signal and a fan rotating speed signal are collected; identifying the fan sound signal and intercepting a stable section sound signal; a weighting processing is carried out on the sound signal in the stable section so as to obtain sound features reflecting loudness perception of human ears; performing short-time Fourier transform on the sound signal subjected to A weighting processing to obtain a time-frequency graph; and inputting the time-frequency diagram through a ResNet18 deep learning model to carry out fan abnormal sound classification so as to obtain an intelligent detection result of the fan operation state. Intelligent detection of the running state of the fan is achieved.

An anti-snoring system comprising an adjustable bed having a sleeping surface that may be mechanically raised or lowered, and a control module adapted to receive commands from a source external to the adjustable bed; a mobile device in direct or indirect communication with the control module in the adjustable bed, and wherein the mobile device has sound recording capabilities; a mobile application resident on the mobile device, wherein the mobile application includes a sound classification machine learning model that includes an artificial intelligence or neural network operative to determine whether or not a person on the sleep surface is snoring, and wherein upon a determination that the person is snoring and has been snoring for a predetermined period of time, the mobile application instructs the control module to raise or adjust the sleeping surface to a height or position that will discourage the person from snoring.

The present application belongs to the technical field of the intelligent interaction of vehicles, and particularly relates to an interaction method, apparatus and system, a sound classification model training method, and a vehicle. The interaction method comprises the following steps: in response to a user operation, calling first scene data, which at least comprises a first visual interaction interface; sending the first scene data to an execution terminal, which at least comprises a display screen; determining whether a blow interaction instruction is detected; when the blow interaction instruction is detected, calling second scene data, which at least comprises a second visual interaction interface; and sending the second scene data to the execution terminal. In the present application, a blow interaction instruction is used, such that a more flexible and more visual human-vehicle interaction experience is realized; a sound classification model and a face recognition model are used, such that the blow interaction instruction can be accurately determined; and different scenes are switched or controlled, such that the technology is of great significance to an improvement in driving safety and user experience.

The application provides a micro-motor abnormal sound classification method and device based on multi-scale feature fusion and attention mechanism, which comprises the following steps: first, sound signal data acquisition; second, sound signal preprocessing; third, sound data feature extraction; fourth, multi-scale feature fusion: a convolutional neural network with three different scale convolution kernels is used to extract multi-level feature information of the comprehensive feature vector to obtain a feature map; a channel attention and a spatial attention mechanism are integrated in each convolutional neural network, the importance of each channel feature vector is weighted through the channel attention mechanism, and the attention area of the feature vector is dynamically adjusted in the spatial dimension through the spatial attention mechanism; fifth, feature classification and judgment of the running state or fault type of the current micro-motor. The method introduces the fusion of multi-scale convolution and attention mechanism, enhances the robustness of the model in a complex environment, and improves the accuracy of fault classification.

The invention discloses a chicken flock sound event detection method based on multi-scale perception and time domain dynamic enhancement, and the method comprises the steps: carrying out the random screening and marking based on chicken flock sound data, and obtaining a standard data set with a label; preprocessing the sound signals in the standard data set to obtain a spectrogram; intercepting a spectrum picture segment containing chicken flock sound in the spectrogram, and converting the spectrum picture segment into a Mel spectrogram through a Mel filter bank; constructing a depth feature extractor, and extracting sound depth features of the Mel spectrogram by using the depth feature extractor; and constructing a frame level sound classifier, processing and analyzing sound depth features, and obtaining time domain distribution of chicken flock sound events. According to the method, the time domain distribution of the sound events of the chicken flocks can be accurately and quickly extracted in a long-time and noisy breeding environment.

The invention relates to the technical field of artificial intelligence, and provides an intelligent abnormal sound sensing power fault diagnosis question-answering system and a method thereof. The method comprises the following steps: acquiring power failure abnormal sound collected by a sound receiver, and identifying the abnormal sound through an identification module; calling a fast question and answer language model to retrieve answers from a vector database according to the identified abnormal sound classification, vector data in the vector database being real number vectors formed according to text fragments related to power grid operation, maintenance and management knowledge, if the answer is retrieved, filling the retrieved answer into a guide word template p1 to generate a standardized answer text, and outputting the standardized answer text to the user; and if no answer is retrieved, calling a deep reasoning language model to call an execution tool related to a power failure reasoning task corresponding to the current question according to the question to perform online answer search, filling a question text and a search result into a guide word template p2 to generate a second standardized answer text, and outputting the second standardized answer text to the user. According to the invention, through abnormal sound identification and cooperative work of the two intelligent models, multi-mode accuracy and reply speed of power failure question answering are considered.

The present invention provides a method and system for constructing a sound classification target model, which relates to the technical field of sound signal processing, and includes: obtaining original sound data and auxiliary modality data, performing sound data conversion processing on the original sound data to obtain corresponding multi-resolution sound image data; processing the auxiliary modality data to obtain corresponding auxiliary modality spatio-temporal matrices; performing local feature processing on the multi-resolution sound image data and the auxiliary modality spatio-temporal matrices to obtain local sound image features and auxiliary modality features; constructing a hierarchical feature sequence space according to the local sound image features and the auxiliary modality features to obtain corresponding fused feature sequences; constructing a sound classification target model according to the fused feature sequences, generating a dynamic sound index according to the sound classification target model, and further obtaining a sound classification result, thereby improving the accuracy of sound classification.

This disclosure provides systems, methods, and devices for audio signal processing that support noise cancellation. In a first aspect, a method of signal processing includes determining a location of the apparatus; receiving an audio signal including sounds at the location of the apparatus; determining, based on a machine learning (ML) model, to reduce a presence of the one or more sounds in the audio signal based on the location; and determining an output audio signal by reducing the presence of the one or more sounds in the audio signal. Other aspects and features are also claimed and described.

The present invention provides a sound classification method and device, a storage medium, and a computer. The sound classification method includes: providing a training set, a validation set, and a convolutional neural network model; obtaining time domain features and frequency domain features from a sample by windowing, superimposing the extracted time domain features and frequency domain features to obtain a time-frequency combination feature of the sample; inputting the time-frequency combination feature obtained based on the training samples in the training set into the convolutional neural network model for training, and inputting the time-frequency combination feature obtained based on the validation samples in the validation set into the convolutional neural network model for verification, and obtaining a trained and verified convolutional neural network model through multiple training and verification cycles. In this way, while taking into account the changes in frequency energy, the overall sound information also has a high degree of recognition, and the accuracy of classification can also be improved.

The application relates to the technical field of machine learning, and provides an animal sound classification method, system, device, medium and product. The application trains a classification model through a staged strategy, including pre-training the classification model based on a general audio dataset, end-to-end training the classification model based on an animal sound dataset, and deploying the trained classification model to an intelligent device; the intelligent device collects original audio signals in real time and extracts log-mel spectrum features as feature inputs of the classification model, and the model finally outputs prediction probabilities for each target sound category, so as to determine an animal sound classification result. The staged training method significantly reduces the model training difficulty, the log-mel spectrum features are used as inputs to improve the robustness of the model to background noise, and the generalization ability and overall accuracy of the classification model are effectively improved.

The application provides an air conditioner production line abnormal sound detection method and system, comprising: acquiring sound signals in real time through a microphone spherical array; processing the sound signals based on an HOA-SHT domain analysis framework to generate a panorama sound image; acquiring an air conditioner image through a panorama camera; locking an air conditioner area according to the air conditioner image; performing multi-modal positioning on the air conditioner area through the panorama sound image to obtain a target area; extracting sound signals from the target area through a multi-scale mel spectrum, and inputting the extracted sound signals into a beamformer of a neural network to obtain noise signals; detecting and identifying the noise signals based on a convolutional neural network-based anomaly classifier to obtain air conditioner abnormal sound classification results; efficiently and accurately detecting air conditioner hanging machine abnormal sounds, effectively overcoming the limitations of traditional manual detection methods, significantly improving production efficiency, reducing the false detection rate and the missed detection rate, and providing strong quality guarantee for the production line.

The application discloses a bird chirp sound recognition method based on a combination of voiceprints and spatial distribution, and belongs to the technical field of intelligent sound classification and recognition. In view of the problems of ignoring geographical distribution prior knowledge and sample imbalance in the prior art, the application firstly constructs a voiceprint recognition model: a training data set is constructed by audio preprocessing, logarithmic mel spectrum and dynamic difference feature extraction, a model is trained based on DenseNet-121 by adopting a two-stage training strategy, and recognition confidence of each species is obtained; meanwhile, a spatial distribution model is constructed: based on public observation data, an average observer ability index is used to correct an original encounter rate, and spatial distribution probability of the species in a specific city is obtained; finally, a Sigmoid function is used to perform nonlinear fusion on the two, a joint recognition probability is calculated, and a classification result is output. The application introduces ecological spatial constraints into the recognition decision, effectively reduces false positive misjudgment, improves rare species monitoring capability, and makes the recognition result have ecological interpretability.

A system method and computer program product for hierarchical classification of sounds includes one or more neural networks implemented on one or more processors. The one or more neural networks are configured to classify sounds into two or more hierarchical procedural and fine-grained levels of classification in a hierarchy. The classified sounds can be used to search a database for similar or contextually related sounds.

The invention discloses a sound classification method, device and equipment based on improved African vulture, and a storage medium, and relates to the technical field of sound classification, and the method comprises the steps: extracting the Mel frequency cepstral coefficient features of sound data; optimizing the African vulture optimization algorithm according to a preset optimization strategy to obtain a target African vulture optimization algorithm, wherein the preset optimization strategy comprises a Cauchy Gaussian mixture variation strategy, an adaptive weight coefficient strategy and a simplex method strategy; optimizing hyper-parameters of a long and short term memory network model according to the target African vulture optimization algorithm to obtain a target hyper-parameter combination; and reconstructing the long and short-term memory model according to the target hyper-parameter combination and the Mel-frequency cepstral coefficient features to obtain a sound classification model so as to realize sound classification. According to the method, the new sound classification model is constructed by improving the African vulture optimization algorithm, so that the accuracy and the stability of urban sound classification are improved.

The present application relates to the technical field of musical instrument tuning, and particularly relates to a self-learning musical instrument pitch correction device capable of eliminating environmental noise. The present application comprises an image acquisition module, a vibration feature acquisition module, a fundamental frequency acquisition module, a correction indication module and a self-learning recording module. The image acquisition module is used to collect string vibration images to physically isolate noise, the vibration feature acquisition module is used to track string displacement to generate vibration feature data, the fundamental frequency acquisition module is used to perform frequency domain transformation and interpolation on the data to obtain the actual vibration fundamental frequency, the correction indication module is used to compare with the standard pitch and output a graded color indication and a tuning direction, and the self-learning recording module is used to analyze previous deviation data to generate stability evaluation and maintenance suggestions. The present application realizes sound classification level pitch detection and correction without sound collection, and improves the tuning reliability in a noisy environment.

The invention relates to the field of computer technology and audio signal processing, and provides a brain-like low-power-consumption audio classification method and device suitable for edge equipment, equipment, a medium and a product. Decomposing the target audio by using an audio decomposition module of the audio classification model to obtain a first sub-band and a second sub-band; performing feature extraction on the first sub-band and the second sub-band by using a feature extraction module of the audio classification model to obtain a first feature corresponding to the first sub-band and a second feature corresponding to the second sub-band; and predicting a sound classification of the target audio based on the first feature and the second feature by using a classification module of the audio classification model. According to the method and the device, the problem of redundant calculation caused by homogenization processing of the audio can be solved, the problem that high-frequency information and low-frequency information cannot be distinguished is avoided, targeted feature extraction can be performed on different frequency components, so that redundant calculation can be avoided, and the calculation efficiency of audio classification is improved.

The present application discloses a training method, a sound classification method, a device and equipment for an improved prototype network, wherein the improved prototype network is composed of a time-domain attention convolution module, a frequency-domain attention convolution module, a recurrent neural network layer and a fully connected layer, and the network model can be trained using a small amount of audio data. When performing sound classification, the above-mentioned trained improved prototype network is used to extract the prototype feature representation of the audio to be tested, and the distance between the prototype feature representation of the audio to be tested and the prototype feature representation of the known audio category is calculated. The matching score of the audio to be tested belonging to the known sound category is calculated based on the above distance value, and finally, the audio to be tested is judged to be a known category or an unknown category based on the matching score. If it is a known category, its specific category is predicted based on the matching score. The present invention realizes the classification of open environment sound under small sample data, which can well solve the problems caused by the small number of collected sound samples and incomplete sound categories in real life.