Online instruction word speech recognition method and system in noise environment

A speech recognition and environmental technology, applied in speech recognition, speech analysis, neural learning methods, etc., can solve the problem of poor speech recognition effect of instruction words, improve the recognition accuracy, solve the problem of open set recognition, and achieve the effect of accurate recognition

Pending Publication Date: 2022-01-11
HARBIN INST OF TECH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of the above problems, the present invention proposes an online instruction word speech recognition method and system in a noisy en...

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
  • Online instruction word speech recognition method and system in noise environment
  • Online instruction word speech recognition method and system in noise environment
  • Online instruction word speech recognition method and system in noise environment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0044] In order to enable those skilled in the art to better understand the solutions of the present invention, exemplary implementations or embodiments of the present invention will be described below in conjunction with the accompanying drawings. Apparently, the described embodiments or examples are only part of the embodiments or embodiments of the present invention, not all of them. Based on the implementation modes or examples in the present invention, all other implementation modes or examples obtained by persons of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0045] CNN does not rely on the idea of ​​​​sequence reasoning, and has a certain tolerance to noise. Based on this characteristic, the present invention uses the CNN classification model to convert the command word speech recognition problem into an image recognition problem for processing, which effectively solves the disadvantages of the ge...

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 an online instruction word speech recognition method and system in a noise environment, belongs to the technical field of speech recognition, and is used for solving the problem that an existing sequence reasoning model is poor in instruction word speech recognition effect in the noise environment. According to the technical key points, the method comprises the following steps: a voice recognition problem is converted into an image recognition problem for processing by using a CNN classification model, and voice and noise are accurately distinguished by using a CNN dichotomy network model; furthermore, a classification judgment method based on activation vector input is provided, the activation vector of a CNN network model is used as a judgment basis, classification of unknown category voice and instruction word voice is accurately achieved, and the open set recognition problem is well solved. According to the invention, non-instruction word speech can be effectively rejected, and accurate recognition of instruction word speech in a noise environment is realized. The method is suitable for noise-containing instruction word speech recognition when an intelligent mining robot or other equipment works.

Description

technical field [0001] The invention relates to the technical field of speech recognition, in particular to a method and system for speech recognition of online command words in a noisy environment. Background technique [0002] Speech recognition is a technology that enables machines to recognize or understand external voice input. This technology is the basis for further interaction between humans and machines. Only when the machine can recognize the input voice can it further understand the recognized content and make corresponding feedback. At present, the development prospect of speech recognition technology is very broad. As a direction of intelligent mining robot development, it can assist the driver to control the intelligent mining robot, reduce the difficulty of operating the intelligent mining robot, and has a wide application prospect. [0003] The rapid development of deep learning in recent years has attracted many researchers to devote their energy to researc...

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): G10L15/20G10L15/02G10L15/08G10L15/16G10L15/22G10L25/24G10L25/87G10L25/84G06N3/04G06N3/08G06K9/62
CPCG10L15/20G10L15/02G10L15/08G10L15/16G10L15/22G10L25/24G10L25/87G10L25/84G06N3/08G10L2015/223G06N3/047G06N3/045G06F18/2415
Inventor 王波霍鑫吴鑫䶮
Owner HARBIN INST OF 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
Try Eureka
PatSnap group products