Noise classification method of Gaussian Mixture Model based on neural network

A Gaussian mixture model, noise classification technology, applied in the field of speaker recognition

Inactive Publication Date: 2012-09-26
张燕
View PDF4 Cites 44 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, at present, GMM and auto-associative neural network (AANN) are only used for noise classification alone, and there is no method that combines the respective advantages of the two to better improve the effect of noise classification.

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 classification method of Gaussian Mixture Model based on neural network
  • Noise classification method of Gaussian Mixture Model based on neural network
  • Noise classification method of Gaussian Mixture Model based on neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0075] The technical solutions of the present invention will be further described below in conjunction with the drawings and embodiments.

[0076] figure 1 is the noise type training and recognition model, which differs from the baseline GMM model (only GMM model is used as noise recognition) in training and other aspects. figure 2 is the AANN network model.

[0077] (1) Preprocessing and feature extraction;

[0078] First, a method based on energy and zero-crossing rate is used for silence detection, and the noise signal is pre-emphasized, framed, and linear predictive (LPC) analysis is performed, and then the cepstral coefficient is obtained from the obtained LPC coefficient as the noise A feature vector for classification.

[0079] (2) Training;

[0080] During training, the extracted eigenvectors are delayed as the input of AANN, AANN learns the structure of eigenvectors, and extracts the time information of eigenvector sequences. Then the learning results are provid...

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 discloses a noise classification method of Gaussian Mixture Model based on a neural network. The recognition rate of noise classification can be improved by using the method. When the method is training, the extracted eigenvector acts as the input of auto-associative neural network (AANN) after being delayed, the structure of the eigenvector is studied by AANN, and the temporal information of the eigenvector sequence is extracted. Then the study result is provided to Gaussian Mixture Models (GMM) in the form of residual eigenvector. The GMM training is carried out by using Expectation Maximization (EM). The weight coefficient of AANN is updated using the method of backward inversion with inertia. The noise classification method makes full use of the advantages of AANN and GMM and greatly improves the recognition rate of the whole noise classification system.

Description

technical field [0001] The invention relates to a speaker recognition method, in particular to a noise classification method based on a Gaussian mixture model of a neural network. Background technique [0002] Reducing all kinds of urban environmental noise is one of the important indicators of modern urban environmental protection, which directly affects the image of the city. Urban environmental noise monitoring system is an essential public facility in the city. The application of noise monitoring system can improve the hardware level of environmental protection and enhance the reliability and controllability of urban environmental protection. According to the continuous monitoring at different points in the city, the phenomenon of violating the noise decibel can be found in real time, and the work efficiency of environmental protection can be improved. [0003] In the method of noise classification, the method based on Gaussian mixture model (GMM) has been paid more an...

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): G10L15/16G10L15/02G10L15/06G10L17/04
Inventor 张燕姜志鹏姚健东唐加能陈存宝黄艳蔡群李国华
Owner 张燕
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap