Dual-microphone noise reduction method with adjustable expected sound source direction

Abstract

The invention discloses a dual-microphone noise reduction method with an adjustable expected sound source direction, and the method comprises the steps: carrying out the discrete sampling, pre-emphasis, framing and windowing processing of noise signals x1 (t) and x2 (t) received by dual microphones, and obtaining frequency domain signals X1 (omega) and X2 (omega) through short-time Fourier transform; in the beam forming process, a virtual microphone is introduced at the midpoint of a connecting line of the double microphones, differential transformation is carried out on the frequency domain signals X1 (omega) and X2 (omega) according to a central difference format, and differential signals Y1 (omega) and Y2 (omega) are constructed. And calculating statistical average values of power spectrums of the differential signals Y1 (omega) and Y2 (omega), marking a ratio of the statistical average values as a directivity function gamma (omega, theta), analyzing properties of the directivity function gamma (omega, theta), and directly mapping the directivity function gamma (omega, theta) into a noise masking value lambda (omega) through a normalization function. Multiplying the X1 (omega) by the lambda (omega) to obtain a signal R1 (omega) after the noise in the competition direction is eliminated; and a post Wiener filtering process: estimating signal energy and noise energy in the R1 (omega) to obtain a channel signal-to-noise ratio, calculating a gain function, and further eliminating residual noise in the R1 (omega).

Description

technical field [0001] The present invention relates to the technical field of noise reduction of speech signals, and in particular, to a method for noise reduction of dual microphones with adjustable desired sound source direction. Background technique [0002] Portable devices such as Bluetooth headsets have gradually become a good tool for improving efficiency in people's daily lives since their introduction. However, when users use them to make or receive calls, if they are disturbed by background noise and non-target voices, the call quality will drop sharply. In this case, it is an urgent need to retain the voice close to the user's speaking direction, and to suppress background noise and non-target voice as much as possible on the premise of ensuring that the voice is not distorted. [0003] Existing generalized sidelobe cancellers (GSCs) and delayed beamformers use multiple microphone recording signals for spatial filtering. For portable devices such as Bluetooth he...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problems that the non-target direction sound cannot be accurately eliminated and the desired sound source direction cannot be set according to requirements when beamforming is directly applied to a short-distance dual-microphone system, a two-step denoising method based on beamforming and Wiener filtering is proposed.

Images