Voice signal noise reduction method and device

Abstract

The invention discloses a voice signal noise reduction method and device. The method comprises the steps: obtaining a real-time image; analyzing the real-time image, and determining a plurality of sound source directions from a preset image; labeling the sound source directions to distinguish human voice direction information and noise direction information; performing noise reduction on sound data in the real-time image according to the human voice direction information and the noise direction information; and outputting the noise-reduced sound data. Sound data in the real-time image is subjected to noise reduction by utilizing human voice direction information and noise direction information in the real-time image, so that the sound data in multiple directions is provided for noise reduction in the processing process; and the sound data is subjected to noise reduction by utilizing the human voice direction and the noise direction in the real-time image, and the human voice and the noise in the sound data can be separated through the human voice direction and the noise direction, so that the final denoising effect is more obvious and efficient.

Description

technical field [0001] The invention relates to the technical field of speech signal processing, in particular to a speech signal noise reduction method and device. Background technique [0002] At present, with the advancement and development of science and technology, people's quality of life is getting higher and higher. On the basis of enjoying material life, they are pursuing a higher spiritual life. In their spare time, they go to KTV, bars and other entertainment facilities to relax, but Because the sounds in the above-mentioned environments are relatively noisy, there will be a lot of noise in the process of people talking, which will have a great impact on the communication between people, so a noise reduction method has been developed for this problem. The existing The noise reduction method is to collect the pronunciation information of the speaker by determining the position of the speaker, and based on this information, guide the beamforming to perform speech en...

AI Technical Summary

Problems solved by technology

The above-mentioned technical method has the following disadvantages. The actual principle of simply enhancing the human voice is to reduce the interference of noise by improving the human voice. In essence, there is no place for noise, so that the fixed beamforming method cannot obtain satisfactory noise suppression. effect, and the above method does not have a strong voice enhancement effect in a low SNR environment, which affects the user experience

Images