Abnormal sound extraction and recognition method and device based on audio frequency spectrogram
A technology of abnormal sound and audio frequency spectrum, which is applied in the field of abnormal sound extraction and recognition, can solve the problems of low accuracy and large training samples, and achieve the effect of improving accuracy and reducing training samples
Inactive Publication Date: 2022-07-01
湖南工商大学
View PDF0 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The purpose of the present invention is: in order to solve the problem that the accuracy of the audio information extraction and recognition method in the prior art is not high when performing matching, extraction and recognition of speech, and the required training samples are very large, the present invention provides abnormal sound extraction and recognition based on audio spectrogram Method and device to improve the accuracy of audio information extraction and recognition, and reduce training samples
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0065] In order to make the purposes, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments These are some embodiments of the present invention, but not all embodiments. The components of the embodiments of the invention generally described and illustrated in the drawings herein may be arranged and designed in a variety of different configurations.
[0066] Endpoint-based abnormal sound detection algorithm: In the alarm sound recognition system, the endpoint detection algorithm needs to be used to determine the starting point and end point of the abnormal sound, and then only valid sound signals are stored and processed. The most traditional method of endpoint detection is short-term energy and short-term T...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses an abnormal sound extraction and recognition method based on an audio frequency spectrogram, relates to the technical field of abnormal sound extraction and recognition, and is used for solving the problems that in the prior art, an audio frequency information extraction and recognition method is low in accuracy and needs a large number of training samples during voice matching, extraction and recognition. The method comprises the following steps: preprocessing audio data; performing time-frequency conversion on the training sample, and performing time-frequency conversion on the audio signal of the training sample to form a spectrum animation graph; gradient features of the spectrum animation graph are extracted; obtaining a new feature matrix; model construction: constructing an SVM model through a machine learning algorithm; and obtaining a model, taking the new feature matrix as an input, taking the mark of the audio data as an expected output, and training by using an SVM model to obtain an abnormal sound recognition model. According to the invention, an image recognition technology and an audio recognition technology are combined, so that the accuracy of audio information extraction and recognition can be improved, and corresponding training samples are reduced.
Description
technical field [0001] The present invention relates to the technical field of extraction and identification of abnormal sounds, and more particularly to a method and device for extraction and identification of abnormal sounds based on audio spectrograms. Background technique [0002] With the rapid development of the information industry and computer technology, the data volume of multimedia data such as images, videos, and audio has grown rapidly, and has gradually become the main form of information media in the field of information processing. Among them, audio information occupies a very important position. What people face is not the lack of multimedia data, but how to effectively process, deeply analyze and make full use of the vast multimedia big data. The big dataization of audio information, on the one hand, provides conditions for people's needs, but on the other hand This makes it more difficult for people to manage and retrieve these audio information. [0003]...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G10L15/06G10L15/16G10L25/03G10L25/51G06K9/62G06N20/10
CPCG10L15/063G10L15/16G10L25/03G10L25/51G06N20/10G10L2015/0631G06F18/2411
Inventor 谢小良张樊姚欣平张媛媛周晴情晋友迪毕胜男乔玲贺婷婷宋子睿黄楚然
Owner 湖南工商大学