Real-Time Speech Enhancement Using Spectral Subtraction with Minimum Statistics and Spectral Floor

TL;DR

This paper introduces a real-time spectral subtraction method for speech enhancement that uses minimum statistics for noise estimation, avoiding the need for VAD, and applies spectral floor to reduce musical noise, with additional enhancements for residual noise reduction.

Contribution

The paper presents a novel real-time spectral subtraction technique utilizing minimum statistics and spectral floor, improving noise reduction without requiring voice activity detection.

Findings

Effective noise suppression demonstrated in time-frequency plots

Spectral floor reduces musical noise artifacts

Additional enhancements further improve residual noise reduction

Abstract

An initial real-time speech enhancement method is presented to reduce the effects of additive noise. The method operates in the frequency domain and is a form of spectral subtraction. Initially, minimum statistics are used to generate an estimate of the noise signal in the frequency domain. The use of minimum statistics avoids the need for a voice activity detector (VAD) which has proven to be challenging to create. As minimum statistics are used, the noise signal estimate must be multiplied by a scaling factor before subtraction from the noise corrupted speech signal can take place. A spectral floor is applied to the difference to suppress the effects of "musical noise". Finally, a series of further enhancements are considered to reduce the effects of residual noise even further. These methods are compared using time-frequency plots to create the final speech enhancement design