Supercharge Your Innovation With Domain-Expert AI Agents!

SVM secondary classification method for probability likelihood score of GMM-HMM hybrid model

A hybrid model, secondary classification technology, applied in speech analysis, character and pattern recognition, instruments, etc., to achieve the effect of improving classification accuracy and improving classification accuracy

Inactive Publication Date: 2021-08-10
HARBIN ENG UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] In the field of underwater acoustics based on the GMM-HMM (Gaussian Mixed Model, GMM) hybrid model, there are few such studies in China

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
  • SVM secondary classification method for probability likelihood score of GMM-HMM hybrid model
  • SVM secondary classification method for probability likelihood score of GMM-HMM hybrid model
  • SVM secondary classification method for probability likelihood score of GMM-HMM hybrid model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0025] Hidden Markov is a dynamic pattern analysis and processing method with powerful performance. This method has been widely used in the field of speech recognition and natural language processing, and has achieved amazing achievements. In essence, the HMM acoustic model belongs to a method of pattern recognition, and the classification of whale calls is also a kind of pattern recognition problem.

[0026] In fact, a whale call is an indeterminate vibration signal that changes over time, so its characteristics can be described by multiple dynamic variables. Moreover, since the type of whale is an implicit state value, it cannot be obtained directly by observing the whale call, so it needs to be obtained by manually extracting the characteristics of the whale call as the observation value. Further considering the similarity between...

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 an SVM secondary classification method for probability likelihood scores of a GMM-HMM hybrid model, and the method comprises the steps: carrying out the modeling of a whale sound through a GMM-HMM hybrid model, and carrying out the secondary classification of a model through an SVM. Experimental data processing shows that the method effectively improves the classification accuracy, the classification accuracy of the method is improved by nearly 2% compared with that of GMM-HMM, and the classification accuracy of the method is improved by nearly 10% compared with that of an SVM model.

Description

technical field [0001] The invention relates to an SVM secondary classification method for the probability likelihood score of a GMM-HMM mixed model, and belongs to the technical field of underwater acoustic application of a neural network method. Background technique [0002] A Markov chain describes a Markov process with a discrete, finite state space, or discrete index set. The Markov process corresponds to such a random process in practice, that is, based on the known current state, the evolution of the process itself has nothing to do with the past state. In other words, it is a process that can predict future results only by relying on the current state . In this state of complete dependence on "now", the characteristic of "future" and "past" assuming complete independence is called Markov, and the process with Markov is also called Markov process. [0003] The hidden Markov model is extended on the basis of the Markov chain. Considering that the actual situation is...

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): G06K9/62G10L15/14
CPCG10L15/144G06F18/2411G06F18/295
Inventor 李秀坤王集刘开金
Owner HARBIN ENG UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com