Single Channel Speech Enhancement Using Outlier Detection

TL;DR

This paper introduces a dictionary-based single-channel speech enhancement method that uses outlier detection to effectively reduce noise while preserving speech, especially in non-stationary noisy environments.

Contribution

The proposed method uniquely employs outlier detection on dictionary-matched spectral patches to estimate noise, avoiding the need for a separate noise model.

Findings

Significant noise reduction in non-stationary noise conditions

Effective preservation of underlying speech quality

Outperforms some existing methods in challenging environments

Abstract

Distortion of the underlying speech is a common problem for single-channel speech enhancement algorithms, and hinders such methods from being used more extensively. A dictionary based speech enhancement method that emphasizes preserving the underlying speech is proposed. Spectral patches of clean speech are sampled and clustered to train a dictionary. Given a noisy speech spectral patch, the best matching dictionary entry is selected and used to estimate the noise power at each time-frequency bin. The noise estimation step is formulated as an outlier detection problem, where the noise at each bin is assumed present only if it is an outlier to the corresponding bin of the best matching dictionary entry. This framework assigns higher priority in removing spectral elements that strongly deviate from a typical spoken unit stored in the trained dictionary. Even without the aid of a separate noise model, this method can achieve significant noise reduction for various non-stationary noises, while effectively preserving the underlying speech in more challenging noisy environments.