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

Speech separation method and device

A voice separation and voice signal technology, applied in the field of signal processing, can solve the problems of voice energy holes in voice signals, distorted voice signals, etc., and achieve the effect of reducing energy holes and suppressing distortion

Active Publication Date: 2019-02-01
HUAWEI TECH CO LTD +1
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the wrong estimation of the threshold-based IBM matrix, some speech-dominated time-frequency units will be wrongly rejected, and some noise-dominated time-frequency units will be wrongly reserved, resulting in a lot of speech energy in the separated speech signal. Hollow, thus distorting the original speech signal to a great extent

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
  • Speech separation method and device
  • Speech separation method and device
  • Speech separation method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0055] figure 1 It is a flow chart of the speech separation method provided by Embodiment 1 of the present invention. The method of this embodiment is applicable to the situation of reducing the distortion of the target-separated speech signal based on the computational auditory scene analysis. The method is executed by a speech separation devic...

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 embodiment of the invention provides a voice separation method and device. The voice separation method comprises steps of: acquiring a first signal; determining an initial ideal binary mask (IBM) matrix according to the acquired first signal; performing harmonic compensation on the first signal according to the initial IBM matrix in order to obtain a separate voice signal having been subjected to harmonic compensation; filtering the first signal and a second signal according to the separate voice signal having been subjected to harmonic compensation in order to obtain a target separate voice signal, thereby reducing energy holes generated in the target separate voice signal and suppressing the distortion of the target separate voice signal.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of signal processing, and in particular, to a method and device for voice separation. Background technique [0002] Speech signal processing, as an attractive research field in recent years, has achieved a series of remarkable achievements in large vocabulary continuous speech recognition, speech synthesis, speech communication and so on. However, many of the existing speech signal processing technologies are developed in pure speech or speech environments with weak noise, and satisfactory results cannot always be obtained in noisy environments. The application of some voice-related products in real life is limited. Therefore, how to suppress or eliminate the background noise to separate the target speech signal has become an important research direction in the field of speech signal processing. [0003] Computational auditory scene analysis is mainly based on the research in the fiel...

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): G10L21/0272G10L21/02G10L21/0208
Inventor 杨小洪肖玮梁山刘文举
Owner HUAWEI TECH CO LTD