Blind Mask to Improve Intelligibility of Non-Stationary Noisy Speech

TL;DR

This paper introduces a blind acoustic mask that adaptively detects noise in non-stationary noisy speech, improving intelligibility without prior knowledge of noise or speech statistics, and performs comparably to ideal masks.

Contribution

The proposed blind acoustic mask uses noise statistics to adaptively select speech segments, enhancing intelligibility without requiring prior noise or speech information.

Findings

Achieves intelligibility gains comparable to ideal masks

Maintains good speech quality in non-stationary noise conditions

Effective across various noise types and SNR levels

Abstract

This letter proposes a novel blind acoustic mask (BAM) designed to adaptively detect noise components and preserve target speech segments in time-domain. A robust standard deviation estimator is applied to the non-stationary noisy speech to identify noise masking elements. The main contribution of the proposed solution is the use of this noise statistics to derive an adaptive information to define and select samples with lower noise proportion. Thus, preserving speech intelligibility. Additionally, no information of the target speech and noise signals statistics is previously required to this non-ideal mask. The BAM and three competitive methods, Ideal Binary Mask (IBM), Target Binary Mask (TBM), and Non-stationary Noise Estimation for Speech Enhancement (NNESE), are evaluated considering speech signals corrupted by three non-stationary acoustic noises and six values of signal-to-noise ratio (SNR). Results demonstrate that the BAM technique achieves intelligibility gains comparable to ideal masks while maintaining good speech quality.