Dual-microphone directional noise inhibition system and method fused with fundamental frequency information

Abstract

The invention discloses a dual-microphone directional noise inhibition system and method fused with fundamental frequency information. The system comprises a main sound signal input device, an auxiliary sound signal input device, a single-channel voice signal separating module and a dual-channel voice signal separating module. The method comprises the steps that the main sound signal input device collects main voices and inputs the main voices into the single-channel voice signal separating module, and the auxiliary sound signal input device collects background noise signals; and output signals of the single-channel voice signal separating module and the auxiliary sound signal input device are all input into the dual-channel voice signal separating module, and finally processed sound signals are output. According to the invention, a contradiction problem between noise remaining and voice distortion is fully considered, noises can be effectively inhibited on the premise that voices are not distorted, and most non-stationary noises in the real environment can be processed.

Description

technical field [0001] The present invention relates to a system and method for noise suppression, in particular to a system and method for dual-microphone directional noise suppression by fusing fundamental frequency information. Background technique [0002] Voice-based human-computer interaction technology is the most convenient and natural way of interaction between people and computers, and between people and computers. With the development of smart phones, voice recognition and other related product technologies, the market demand for voice noise reduction related technologies is becoming more and more urgent. When users use smart voice-related products, the target voice is usually accompanied by some noise, such as the sound of TV and audio equipment in the home environment. These noises limit the effect of smart voice and user experience. The technology of this patent invention is based on this requirement. Through signal processing, it can effectively filter out re...

AI Technical Summary

Problems solved by technology

[0005] The speech enhancement technology represented by spectral subtraction has strict requirements on the stability of the noise, and can only effectively deal with white noise or other noise signals with slow energy changes; the speech signal is usually subtracted by mistake, and the speech is greatly distorted , the speech perception quality is poor after noise reduction, and the intelligibility is not improved, and the speech recognition rate will further decrease due to speech distortion

[0006] The microphone array processing technology represented by beamforming requires 4 or more microphones to achieve a relatively good noise reduction effect. It needs to know the prior information of the user's orientation, and has limited performance for directional noise suppression.

[0007] Blind source separation technology represented by principal component analysis works well in the ideal situation of no reverberation or weak reverberation, but in real home environment applications, reverberation is usually very strong, resulting in performance degradation; usually the sound source The number cannot be more than the number of microphones. For example, taking dual microphones as an example, if there is more than one noise source (the user's sound source should also be counted), it will lead to performance degradation; in addition, excessive time overhead is also an important reason for limiting its application.

Images