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

Noise estimation method and device

A noise estimation and variance technology, applied in the field of noise estimation methods and devices, can solve problems such as easy underestimation of noise, easy overestimation of noise, large tracking delay, etc., and achieve the effect of solving tracking delay

Inactive Publication Date: 2019-08-16
AISPEECH CO LTD
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The MCRA algorithm uses a fixed time window when tracking the minimum value of the smooth power spectrum of the noisy speech. Therefore, it is easy to overestimate the noise when the speech frame is continuous, and it is easy to underestimate the noise when the noise increases, resulting in a relatively large noise loss in the case of sudden noise changes. large tracking delay

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
  • Noise estimation method and device
  • Noise estimation method and device
  • Noise estimation method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0029] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments It is a part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

[0030] It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0031] This application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program mod...

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 noise estimation method and device. The method comprises the steps that multiple audio variances of at least one frame signal at multiple frequency points are obtained through calculation according to frequency domain audio signals; the average audio variance corresponding to at least one frame signal is obtained through calculation according to the audio variances; a sigmoid function is adopted, noise estimation is conducted according to the average audio variance. Accordingly, by means of the difference of frequency spectrum average variance distribution of pure noise audio signals and pure voice audio signals, noise frames and voice frames are distinguished, so that the large tracking time delay problem under the noise mutation condition is well solved.

Description

technical field [0001] The present application relates to the technical field of speech signal processing, in particular to a noise estimation method and device. Background technique [0002] Speech recognition technology is getting popularized and applied more and more, for example, various voice assistants are produced. One factor closely related to the accuracy of speech recognition is the noise reduction algorithm. The basic principle of the noise reduction algorithm can be divided into two parts as a whole. The first part, noise estimation, first estimates the power spectrum characteristics of the noise. The second part, speech enhancement, uses the estimated noise PSD to calculate the multiplicative spectral gain, and then estimates the clean speech from the noisy speech. Therefore, the accuracy of noise estimation in the noise reduction algorithm is particularly critical. However, the MCRA algorithm has a long delay in the estimated noise spectrum in the case of su...

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): G10L21/0216G10L21/0232
CPCG10L21/0216G10L21/0232
Inventor 袁斌
Owner AISPEECH CO LTD