Signal characteristic extracting methods for automatic classification of voice, music and noise

A technology of automatic classification and signal characteristics, used in speech analysis, speech recognition, instruments, etc.

Inactive Publication Date: 2009-08-26
杨夙
View PDF1 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The object of the present invention is to provide a group of signal feature extraction methods for voice, music, noise automatic classification, on the basis of the signal feature extraction method proposed in the present

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
  • Signal characteristic extracting methods for automatic classification of voice, music and noise
  • Signal characteristic extracting methods for automatic classification of voice, music and noise
  • Signal characteristic extracting methods for automatic classification of voice, music and noise

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0062] Step 1: Collect a 0.5-second sound signal at a sampling frequency of 16000Hz to obtain a time series [s 1 ,s 2 ,...,s N ], where N=8000;

[0063] Step 2: For time series [s 1 ,s 2 ,...,s N ], calculate the basic features based on fractal Brownian motion (σ 1 , σ 2 ,...,σ N-2 ),here σ k = 1 N - k Σ i = 1 N - k [ s i + k - s i - 1 N ...

Embodiment 2

[0067] Step 1: Collect a 0.5-second sound signal at a sampling frequency of 16000Hz to obtain a time series [s 1 ,s 2 ,...,s N ], where N=8000;

[0068] Step 2: Exactly the same as Step 2 of Example 1, obtain the basic feature (σ 1 , σ 2 ,...,σ N-2 );

[0069] Step 3: For the basic features f(σ 1 , σ 2 ,...,σ N-2 ) to get f(σ 1 , σ 2 ,...,σ N-2 )=[log(A 1 σ 1 +B 1 ), log(A 2 σ 2 +B 2 ),...,log(A N-2 σ N-2 +B N-2 )], here let A 1 =A 2 =...A 100 = 1, A 101 =A 102 =...A 7998 = 0, B 1 =B 2 =...B 7998 =0, then f(σ 1 , σ 2 ,...,σ N-2 )=[log(σ 1 ), log(σ 2 ),...,log(σ 100 )], put f(σ 1 , σ 2 ,...,σ N-2 )=[log(σ 1 ), log(σ 2 ),...,log(σ 100 )] as used for time series [s 1 ,s 2 ,...,s N ] to classify features;

[0070] Step 4: Take the feature vector [log(σ 1 ), log(σ 2 ),...,log(σ 100 )] input parameter adjusted support vector machine classifier, the rest are the same as step 4 of embodiment 1.

Embodiment 3

[0072] Step 1: Collect a 0.5-second sound signal at a sampling frequency of 16000Hz to obtain a time series [s 1 ,s 2 ,...,s N ], where N=8000;

[0073] Step 2: For time series [s 1 ,s 2 ,...,s N ], calculate the features based on the blanket coverage dimension, the specific steps are:

[0074] (1) For a cycle where i is equal to 1 to N, let U i 0 = L i 0 = s i C , Here take C=10000;

[0075] (2) For r=1, 2, ..., R and i = 2, 3, ..., N-1, calculate U i r = max { U i - 1 r - 1 , U i ...

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 belongs to the technical field of voice signal processing, in particular to signal characteristic extracting methods for the automatic classification of voice, music and noise. On the basis of the signal characteristic extracting methods, a voice signal automatically classifying system can be constructed to determine a voice signal is voice, music or noise; and the application fields of the automatic classification of voice, music and noise comprise voice activity detection of a digital communication system and ambience identification of an audiphones. The invention provides three fractal measurement-based voice signal characteristic extraction methods.

Description

technical field [0001] The invention belongs to the field of sound signal processing, and is specifically a group of signal feature extraction methods. On the basis of the method of the present invention, an automatic sound signal classification system can be constructed to automatically identify whether each sound signal collected is voice, music, or noise. Application His research interests include voice activity detection technology in digital communication systems and environmental sound recognition technology in hearing aids. Background technique [0002] Voice activity detection is a pre-processing process of speech coding, which is widely used in modern digital communication systems. The purpose of voice activity detection is to identify whether each frame of sound signal collected is voice, music, or noise. The meaning of voice activity detection is as follows: When When one of the two communication parties is speaking, the other party is generally listening, but the...

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
IPC IPC(8): G10L11/00G10L11/02G10L19/00H04R25/00G10L15/02G10L25/03
Inventor 杨夙
Owner 杨夙
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