Voice enhancement method based on continuous noise estimation

Abstract

The invention discloses a voice enhancement method based on continuous noise estimation. In a logarithmic spectrum domain, a voice model which is trained in advance is used to carry out continuous estimation on a parameter of a background noise, and an estimated noise mean value is used to recover a clean voice. Firstly, acoustic pretreatment and fast Fourier transform are performed on an input voice so as to acquire an amplitude and a phase position of each frame of digital voice; and the amplitude is used for noise estimation and amplitude spectrum subtraction and the phase position is used for recovering a time domain signal. And then, sub-band filtering is performed on an amplitude spectrum of the digital voice and a logarithm is taken to operate so as to acquire a logarithm spectrum, and the logarithmic spectrum domain voice model which is trained in advance is used to extract a noise parameter from a logarithm spectrum characteristic vector of the voice containing the noise in real time. Finally, an estimated noise parameter is used to carry out weighted amplitude spectrum subtraction on the voice containing the noise, and inverse Fourier transform and overlap add are performed on an amplitude of an enhanced voice and a phase position of the voice containing the noise so as to acquire the enhanced voice. In the invention, continuous estimation is performed on a noise parameter according to frames in the voice containing the noise and noise changes are tracked in real time.

Description

technical field [0001] The invention relates to a speech enhancement method for continuously estimating parameters of background noise with a pre-trained speech model in a logarithmic spectrum domain, and recovering pure speech by using the estimated noise mean value, belonging to the technical field of speech signal processing. Background technique [0002] In speech communication, input speech is usually disturbed by background noise, so it is necessary to use speech enhancement algorithm to suppress noise interference, recover pure speech from noisy speech as much as possible, and increase speech intelligibility. [0003] In speech enhancement, it is usually necessary to use an endpoint detection algorithm to judge the starting point and end point of a speech segment, so as to divide the noisy speech into a speech segment and a noise segment. In the noise segment, the mean value of the background noise is estimated by using the pure noise spectrum without speech; in each ...

AI Technical Summary

Problems solved by technology

Although this method can estimate noise parameters in the speech segment, there is also a long delay, that is, after the type or intensity of the noise changes, it usually takes 2 to 3 seconds to detect the change of the noise and obtain new noise parameters.

Images