Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Noise classifying method based on convolutional neural network

A convolutional neural network and noise classification technology, applied in the field of noise classification, can solve the problems of low noise recognition accuracy, poor statistical characteristics, and low noise classification accuracy, so as to reduce computational complexity, improve robustness, and improve The effect of accuracy

Inactive Publication Date: 2019-08-23
TIANJIN UNIV
View PDF6 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the noise classification algorithm currently proposed is not very accurate in noise classification, especially for non-stationary noise, and the recognition accuracy of such noise with poor statistical characteristics is low

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Noise classifying method based on convolutional neural network
  • Noise classifying method based on convolutional neural network
  • Noise classifying method based on convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

specific example

[0063] Step 1: Preprocessing

[0064] The 15 types of noise signals in the Noisex-92 noise library are divided into frames and windowed. The frame length of each frame signal is 512 sampling points, and the window function uses Hamming window. Each continuous 12-frame signal is regarded as a noise sample, and 60,000 samples are selected as a training set, and 10,000 samples are used as a test set.

[0065] Step 2: Extract features

[0066] The 24-dimensional MFCC and ΔMFCC features are extracted from each frame signal in the sample, and the eigenvalues ​​of a total of 12 frames of signals in each sample are combined into a two-dimensional matrix with a size of 12*24 as the time-frequency feature of each sample.

[0067] Step 3: Build the CNN structure

[0068] CNN network consists of input layer 1, first convolutional layer 2.1, first pooling layer 2.2, second convolutional layer 2.1, second pooling layer 2.2, fully connected layer 3 and output layer 4. Such as figure 1 sh...

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

No PUM Login to View More

Abstract

The invention provides a noise classifying method based on a convolutional neural network. The noise classifying method comprises the steps that input noise signals are subjected to framing and windowing, and the window length is 10 ms-30 ms; frequency domain characteristics and time domain characteristics are extracted from each frame of noise signals subjected to framing and windowing, and thusa two-dimensional matrix with the size being 12*24 is constituted; the convolutional neural network is established and constituted by an input layer, a hidden layer, a full-connection layer and an output layer, wherein the hidden layer is constituted by one or more convolution layers and one or more pooling layers which are arranged at intervals; the convolutional neural network is trained; and the frequency domain characteristics and time domain characteristics of each frame of noise signals are input into the trained convolutional neural network to obtain a classifying result. According to the noise classifying method based on the convolutional neural network, the accuracy rate of noise classifying can be effectively increased, the size of the input characteristics is only 12*24, and thecomputation complexity of the convolutional neural network is effectively lowered.

Description

technical field [0001] The invention relates to a noise classification method. In particular, it relates to a noise classification method based on convolutional neural networks. Background technique [0002] There are many types of noise, and their characteristics are also different. According to the characteristics of noise, noise can be divided into the following categories: impulse noise, periodic noise, broadband noise, voice interference, background noise and transmission noise. Studies have shown that in order to obtain better speech enhancement, recognition, and coding effects, it is first necessary to distinguish which type of noise the speech signal is polluted by, and then adopt different solutions. Therefore, noise classification algorithms are essential. [0003] At present, there are many noise classification algorithms, and the key to improving the classification accuracy is two technologies: one is to extract the characteristics of the noise, and the other i...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G10L25/30G10L25/18G10L25/03G10L21/0208
CPCG10L21/0208G10L25/03G10L25/18G10L25/30
Inventor 张涛刘阳
Owner TIANJIN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com