Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Voice recognition device and method for large-scale words

A technology of speech recognition and speech samples, applied in speech recognition, speech analysis, instruments, etc., can solve problems such as non-negligible waiting time

Inactive Publication Date: 2003-07-16
THOMSON LICENSING SA
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0020] Furthermore, another disadvantage of the prior art is that there is a non-negligible waiting time between when the user speaks and when his words are recognized

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
  • Voice recognition device and method for large-scale words
  • Voice recognition device and method for large-scale words
  • Voice recognition device and method for large-scale words

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0047] Compared with the prior art, the basic principle of the present invention is to achieve close cooperation between the speech processing module and the comprehension module by effectively sharing the common part, that is, the language model.

[0048] The representation of the language model must be exploited efficiently according to its usage scenario.

[0049] According to the invention, a single representation of the grammar is used. (According to the prior art, however, the grammar is represented twice: the first time for a language module, e.g., usually in the form of a finite state automaton, and the second time in a syntactic analyzer, e.g., in the parser LL(k ) form. These two modules now carry the same information, the permitted syntactic strings, in two different forms.

[0050] Furthermore, according to the invention, there is no syntactic analysis phase (or "parsing" phase): no sentences are exchanged between the two modules for analysis. "Backtracking" (or ...

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 relates to a voice recognition process comprising a step (601) of acoustic processing of voice samples (201) and a step (602) of determination of a command intended to be applied to at least one device, and in that said steps of acoustic processing and of determination of a command use a single representation (309) in memory (305) of a language model. The invention also relates to corresponding devices (102) and computer program products.

Description

technical field [0001] The invention relates to the field of speech recognition. [0002] In particular, the present invention relates to large-scale lexical speech interfaces. It finds particular application in command and control systems such as television and multimedia. Background technique [0003] Information or control systems are increasingly using voice interfaces for quick and direct interaction with users. As these systems become more complex, the dialogue styles supported must become richer, one of which is entering the realm of large-scale lexical continuous speech recognition. [0004] It is well known that the design of continuous speech recognition systems for large vocabularies requires the generation of language models that define or estimate receivable strings of words that constitute sentences recognized by the language model. [0005] In a large-scale vocabulary system, a language model enables the speech processing module to construct sentences (ie, ...

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): G10L15/14G10L15/18G10L15/193G10L15/22
CPCG10L15/193G10L15/1815G10L15/22G10L2015/223
Inventor 瑟奇·勒胡伊特欧斯弗雷德里克·苏夫莱
Owner THOMSON LICENSING SA
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products