Learning static spectral weightings for speech intelligibility enhancement in noise

Tang, Y; Cooke, M

Abstract

Near-end speech enhancement works by modifying speech prior to presentation in a noisy environment, typically operating under a constraint of limited or no increase in speech level. One issue is the extent to which near-end enhancement techniques require detailed estimates of the masking environment to function effectively. The current study investigated speech modification strategies based on reallocating energy statically across the spectrum using masker-specific spectral weightings. Weighting patterns were learned offline by maximising a glimpse-based objective intelligibility metric. Keyword scores in sentences in the presence of stationary and fluctuating maskers increased, in some cases by very substantial amounts, following the application of masker- and SNR-specific spectral weighting. A second experiment using generic masker-independent spectral weightings that boosted all frequencies above 1 kHz also led to significant gains in most conditions. These findings indicate that energy-neutral spectral weighting is a highly-effective near-end speech enhancement approach that places minimal demands on detailed masker estimation.

Citation

Tang, Y., & Cooke, M. (2018). Learning static spectral weightings for speech intelligibility enhancement in noise. Computer Speech and Language, 49, 1-16. https://doi.org/10.1016/j.csl.2017.10.003

• Journal Article Type: Article
• Acceptance Date: Oct 16, 2017
• Online Publication Date: Nov 8, 2017
• Publication Date: May 1, 2018
• Deposit Date: Dec 18, 2017
• Publicly Available Date: Nov 8, 2019
• Journal: Computer Speech & Language
• Print ISSN: 0885-2308
• Publisher: Elsevier
• Volume: 49
• Pages: 1-16
• DOI: https://doi.org/10.1016/j.csl.2017.10.003
• Publisher URL: http://dx.doi.org/10.1016/j.csl.2017.10.003
• Related Public URLs: https://www.journals.elsevier.com/computer-speech-and-language
• Additional Information: Funders : European Commission; Projects : LISTA; Grant Number: 256230