Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Men and women speaking voice classification method based on SVM and HMM mixing model

A technology of hybrid model and classification method, applied in speech analysis, instruments, etc., can solve the problems of long training time, weak classification ability, and low recognition rate.

Inactive Publication Date: 2014-06-18
SHANGHAI 8D WORLD NETWORK SCI & TECH
View PDF0 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Face-based gender recognition has a flaw: when the lighting conditions are not sufficient, the recognition rate of this method will be significantly reduced
HMM and SVM have different focuses. HMM is better at processing continuous signals, but its ability to distinguish categories is weaker; SVM is more suitable for classification, but it has the disadvantages of only being suitable for a small number of samples and long training time.

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
  • Men and women speaking voice classification method based on SVM and HMM mixing model
  • Men and women speaking voice classification method based on SVM and HMM mixing model
  • Men and women speaking voice classification method based on SVM and HMM mixing model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The present invention will be further described in detail below in conjunction with examples, but the embodiments of the present invention are not limited thereto.

[0036] figure 1 A schematic diagram of the overall framework of the system showing the method for classifying male and female speech voices based on the SVM and HMM hybrid model according to an embodiment of the present invention; figure 1 An embodiment of a method for classifying male and female speech voices based on a hybrid model of SVM and HMM is shown, including a training phase and a recognition phase.

[0037] figure 2 Shows a schematic diagram of the system training process of the male and female speech voice classification method based on the SVM and HMM hybrid model according to an embodiment of the present invention; figure 2 Shown:

[0038] 1) Training stage

[0039] In the first step, the corresponding gene frequency samples are collected, and the sample features are extracted as the inp...

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

This invention discloses a men and women speaking voice classification method based on an SVM and HMM mixing model, belonging to the voice audio processing field. The method uses the SVM and HMM mixing model to classify the voice of the men and women and comprises steps of inputting audio signals of people of different genders, picking up gene frequencies in the voice signals of the speaker, obtaining MFCC characteristics of different gene frequencies, utilizing the SVM and HMM algorithm learnt by the machine to obtain a mixing classification model of different gene frequencies of men and women and finally achieving a goal of distinguishing the gender of a speaker. By utilizing the method, a user can easily obtain the gender of a speaker whose voice is in a certain audio segment, and the method can help the user obtain the gender characteristics of the audio and provide convenience for further audio processing.

Description

technical field [0001] The method belongs to the technical field of speech and audio processing, and specifically relates to a method for classifying male and female speech sounds based on a mixed model of SVM and HMM. Background technique [0002] In recent years, machine recognition of human gender has become a research hotspot, using many methods, such as face, voice and so on. Face-based gender recognition has a flaw: when the lighting conditions are insufficient, the recognition rate of this method will be significantly reduced. [0003] Speech is the most natural and effective communication method for human beings. Speech gender recognition involves a wide range of fields including: signal processing, artificial intelligence, pattern recognition, acoustics, linguistics, electronic technology, communication theory, neurophysiology and neuropsychology and many other disciplines. How to let the machine recognize the gender of a person, so as to make a more intelligent r...

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): G10L17/14
Inventor 王雷
Owner SHANGHAI 8D WORLD NETWORK SCI & TECH
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
Eureka Blog
Learn More
PatSnap group products