Adaptive context for automatic speech recognition systems

A technology of speech recognition and speech recognition engine, applied in the field of communication systems, can solve the problem that the system cannot effectively adaptively adjust the context information.

Inactive Publication Date: 2008-05-21
QNX SOFTWARE SYST (WAVEMAKERS) INC (CA) +1
View PDF0 Cites 201 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Despite this approach to programming, some systems cannot effectively adapt to available contextual information

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
  • Adaptive context for automatic speech recognition systems
  • Adaptive context for automatic speech recognition systems
  • Adaptive context for automatic speech recognition systems

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0017] After adaptive recognition, the system can adaptively adjust words, phrases and / or sentences. The system can edit the speech recognized from the audio signal, or revise the recognition score associated with the recognized speech. Certain post-recognition systems edit or correct data interactively in real-time or near real-time. Other post-recognition systems edit or correct data through user modification, or a combination of user modification and real-time or near real-time user interaction. The post-recognition system may interface with a speaker-dependent and / or speaker-independent automatic speech recognition system (SRS).

[0018] FIG. 1 is a block diagram of an adaptive automatic speech recognition system 100 . The adaptive automatic speech recognition system 100 may include a speech recognition engine 102 , an adaptive post-recognition system 104 , a translator 106 and a dialog manager 108 . The speech recognition engine 102 receives the digital audio signal an...

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

A system improves speech recognition includes an interface linked to a speech recognition engine. A post-recognition processor coupled to the interface compares recognized speech data generated by the speech recognition engine to contextual information retained in a memory, generates a modified recognized speech data, and transmits the modified recognized speech data to a parsing component.

Description

[0001] priority statement [0002] This application claims priority to US Provisional Application No. 60 / 851,149, filed October 12, 2006, which is incorporated herein by reference. technical field [0003] The present invention relates to communication systems, and more particularly to systems for improving speech recognition. Background technique [0004] Some speech recognition systems interact with applications through exchanges. These systems understand a limited number of spoken requests and commands. Due to the variety of speech patterns, speaker accents, and application environments, some speech recognition systems may not always be able to recognize the user's speech. Some systems try to minimize errors by asking the user to read multiple words and sentences to train the system before using it. Other systems adaptively adjust their speech models during use of the system. Since there are multiple ways in which requests and commands can be obtained, developers of s...

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): G10L15/00G10L15/06G10L15/28
CPCG10L15/26G10L2015/226G10L2015/228G10L15/1822G10L15/06G10L15/28G10L15/22
Inventor R·伦佩尔P·A·赫瑟林顿M·亨内克D·威利特
Owner QNX SOFTWARE SYST (WAVEMAKERS) INC (CA)
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