Unlock instant, AI-driven research and patent intelligence for your innovation.

Selective noise suppression during automatic speech recognition

A technology for automatic speech recognition and noise suppression, applied in speech recognition, speech analysis, instruments, etc., to solve problems such as driver frustration

Active Publication Date: 2016-05-04
GM GLOBAL TECH OPERATIONS LLC
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Of course, in the event the driver tries to command the vehicle using his / her voice and the vehicle doesn't respond (or responds incorrectly), this can lead to driver frustration, and is much more frustrating than if the driver were only manually executing the command without trying to speak Controlling Situations Eventually Lead to Greater Distraction

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
  • Selective noise suppression during automatic speech recognition
  • Selective noise suppression during automatic speech recognition
  • Selective noise suppression during automatic speech recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] The methods described below are generally concerned with improving speech recognition accuracy. Methods describe selectively and intelligently removing noise suppression during front-end or pre-processing of speech signals to achieve this increased accuracy. As will be described in more detail below, automatic speech recognition (ASR) processing may include pre-processing stages and downstream processing stages. A pre-processing stage may typically prepare the received speech signal amidst background noise for interpretation during downstream processing stages. During preprocessing, one or more noise suppression techniques or algorithms may be applied to the speech frames of the speech signal to improve the overall ASR accuracy - thereby providing a cleaner speech signal to downstream processing stages. However, it has been found that in some cases this may introduce undesired musical distortion into downstream stages. These distortions are undesirable because speech ...

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 selective noise suppression during automatic speech recognition. An automatic speech recognition engine and a method of using the engine are described. The method pertains to front-end processing an audio signal and includes the steps of: identifying a plurality of voiced-frames of the audio signal; determining that one or more of the plurality of voiced-frames have a signal-to-noise (SNR) value greater than a first predetermined threshold; and based on the determination, bypassing noise suppression for the one or more of the plurality of voiced-frames.

Description

technical field [0001] The present invention relates to selective noise suppression during automatic speech recognition. Background technique [0002] Hands-free computing devices are becoming increasingly popular; especially in vehicles where voice command / control can be expected - enabling vehicle functions to be controlled using the driver's voice. Voice control systems may receive the driver's voice as input, interpret the voice, and make logical decisions (eg, perform control of vehicle functions) based on the interpretation of the voice. When the driver is able to adopt the voice command mode, the driver's distraction during vehicle operation can be minimized, and distraction-related accidents can be prevented. [0003] Of course, in the event the driver tries to command the vehicle using his / her voice and the vehicle doesn't respond (or responds incorrectly), this can lead to driver frustration, and is much more frustrating than if the driver were only manually execu...

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): G10L15/08G10L15/20G10L21/0208G10L21/0316
CPCG10L15/08G10L15/20G10L21/0208G10L21/0316G10L2015/081G10L25/93
Inventor G.塔尔瓦X.赵R.D.辛斯三世M.F.R.乔扈里
Owner GM GLOBAL TECH OPERATIONS LLC