Unlock instant, AI-driven research and patent intelligence for your innovation.

Method for voice signal feature selection

A voice signal and feature selection technology, which is applied in voice analysis, voice recognition, instruments, etc., can solve the problems of many voice signal image interference items, difficulty in selecting good recognition features, etc., and achieve the effect of improving anti-interference ability and stability

Inactive Publication Date: 2019-11-26
刘小英
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The technical problem to be solved by the present invention is that there are too many interference items in the current voice signal image, and it is difficult to select a good recognition feature. The purpose is to provide a method for voice signal feature selection to solve the above problems

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

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0019] The present invention is used for the method for voice signal feature selection, comprises voice signal sample image, also comprises the following steps:

[0020] S1. After performing image denoising processing, perform image information statistics on the voice signal sample image to obtain a basic statistical feature set;

[0021] S2, obtaining a feature point set based on the image's own attributes;

[0022] S3, acquiring a set of feature points defined by the region;

[0023] S4. Calculate the variance of each feature set sub-item based on the feature set obtained in S1-S3, and select the 6 features with the smallest variance, and then use the feature compression method to reduce the dimension of the 6 features to form 3 description features as an image The features of are used as the feature parameters of the speech signal image.

[0024] The set of feature points defined by the region includes LOG operator, Forstner operator, and SIFT operator.

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 method for voice signal feature selection, and relates to voice signal feature selection. Voice signal sample images are included. The method also comprises the following steps of Step 1, after image denoising processing, performing image information statistics on a voice signal sample image to obtain a basic statistics feature set; Step 2, obtaining a feature point set based on image self attributes; Step 3, obtaining a feature point set defined by a region; and Step 4, calculating a variance of each feature set subitem according to the feature set obtained in the Step 1 to the Step 3, selecting six features with the smallest variances, then, performing dimension reduction on the six features by using a feature compression method, and forming three description features to be used as features of the image and to be used as feature parameters of a voice signal image. By selecting proper features, the stability, precision and real-time performance of the voice signal image feature selection can be effectively improved; the feature effectiveness under different conditions can be comprehensively considered; and the anti-interference capability of the image features is improved.

Description

technical field [0001] The invention relates to speech signal feature selection, in particular to a method for speech signal feature selection. Background technique [0002] Speech recognition has a wide range of applications. The common application systems include: voice input system, which is more in line with people's daily habits, more natural and efficient than keyboard input; voice control system, which uses voice to control equipment. Operation is faster and more convenient than manual control, and can be used in many fields such as industrial control, voice dialing system, smart home appliances, voice-activated smart toys, etc.; the intelligent dialogue query system operates according to the customer's voice, providing users with natural , Friendly database retrieval services, such as family services, hotel services, travel agency service systems, ticket booking systems, medical services, banking services, stock inquiry services, etc. Speech recognition mainly has t...

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): G10L15/02G06K9/62
CPCG10L15/02G06V10/758
Inventor 刘小英
Owner 刘小英