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

Method and device for determining speech recognition threshold in noisy environment

A technology for identifying thresholds and environments, applied in speech analysis, instruments, etc., can solve the problems of tedious and time-consuming determination of speech recognition thresholds, achieve the effects of improving acquisition speed, increasing application range, and reducing operation steps

Active Publication Date: 2019-07-16
SOUTH UNIVERSITY OF SCIENCE AND TECHNOLOGY OF CHINA
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In view of this, the embodiments of the present invention provide a method and device for determining the speech recognition threshold in a noisy environment, so as to solve the technical problems in the prior art that the speech recognition threshold determination process is cumbersome and time-consuming, and the test needs to be re-tested for different noises

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
  • Method and device for determining speech recognition threshold in noisy environment
  • Method and device for determining speech recognition threshold in noisy environment
  • Method and device for determining speech recognition threshold in noisy environment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] Embodiment 1 of the present invention provides a method for determining a speech recognition threshold in a noise environment. The method can be executed by a device for determining a speech recognition threshold in a noise environment, wherein the device can be implemented by software and / or hardware. figure 1 It is a schematic flowchart of a method for determining a speech recognition threshold in a noise environment provided by Embodiment 1 of the present invention. like figure 1 As shown, the determination method includes:

[0032] S110. Acquire speech intelligibility indices corresponding to different signal-to-noise ratio strengths in a noise environment based on the speech intelligibility model.

[0033] In this embodiment, the speech intelligibility model refers to the calculation model for obtaining the speech intelligibility index of the noisy speech signal through the acoustic information of the noisy speech signal; the speech intelligibility index refers t...

Embodiment 2

[0041] figure 2 It is a schematic flowchart of a method for determining the speech recognition threshold in a noisy environment provided by Embodiment 2 of the present invention. This embodiment is optimized on the basis of the above-mentioned embodiments. Further, the speech intelligibility model is used to obtain The speech intelligibility index corresponding to different signal-to-noise ratio strengths, including: dividing the noise-free signal and the noise-bearing signal into a preset number of frequency-domain sub-signals, and obtaining the acoustic information of each frequency-domain sub-signal; calculating the frequency of each noise-free signal Acoustic information difference between domain sub-signal and frequency-domain sub-signal with noise signal; acquire speech intelligibility index under noise environment according to the acoustic information difference.

[0042]Further, said obtaining the signal-to-noise ratio corresponding to the preset speech clarity index ...

Embodiment 3

[0068] image 3 It is a schematic flowchart of a method for determining a speech recognition threshold in a noise environment provided by Embodiment 3 of the present invention. The method for determining the speech recognition threshold in a noise environment can be executed by a device for determining a speech recognition threshold in a noise environment. like image 3 As shown, the determination method includes:

[0069] S310. In a noisy environment, record a section of environmental noise signal, such as Figure 4A As shown, where the abscissa is time, and the ordinate is the signal amplitude of the environmental noise signal.

[0070] S320, a section of existing noise-free speech signal B1 (such as Figure 4B shown) to enhance 5dB and attenuate 5dB respectively to form noise-free speech signal B2 (such as Figure 4C shown) and the noise-free speech signal B3 (such as Figure 4D shown).

[0071] S330. Superimpose the environmental noise signal with the noise-free spee...

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 a method and a device for determining a speech recognition threshold in a noise environment. The method comprises the steps that speech intelligibility indexes corresponding to different signal-to-noise ratios in the noise environment are acquired based on a speech intelligibility model; the signal-to-noise ratio corresponding to the preset speech intelligibility index is acquired according to the correlations between the speech intelligibility indexes and the signal-to-noise ratios; a mathematical model based on the correlation between the signal-to-noise ratio corresponding to the preset speech intelligibility index and the speech recognition threshold is used to determine the speech recognition threshold in the noise environment. By adopting the above mentioned technical scheme, the speech recognition thresholds in different noise environments are determined without carrying out speech recognition tests, and therefore operation steps required by determining the speech recognition thresholds are effectively reduced, the acquisition speeds of the speech recognition thresholds in the noise environment are accelerated, and the application range of the speech recognition threshold determining method is enlarged.

Description

technical field [0001] The invention relates to the technical field of sound processing, in particular to a method and device for determining a speech recognition threshold in a noise environment. Background technique [0002] The noise in the real environment has a great impact on people's speech recognition, and the speech recognition threshold (Speech Reception Threshold) in the noise environment is an important indicator to measure the impact of noise on the speech recognition rate. [0003] The meaning of the speech recognition threshold in a noisy environment is the signal-to-noise ratio when the subject can understand half of the test speech or the speech recognition rate reaches 50%. Traditional methods usually require speech recognition experiments in a noisy environment to measure the speech recognition threshold; or, the intelligibility index of the noisy speech signal is calculated from the noisy speech signal, and the speech intelligibility index (Speech Intelli...

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 Patents(China)
IPC IPC(8): G10L25/48
CPCG10L25/48
Inventor 陈霏
Owner SOUTH UNIVERSITY OF SCIENCE AND TECHNOLOGY OF CHINA