D
Donald S. Williamson
Researcher at Indiana University
Publications - 41
Citations - 1404
Donald S. Williamson is an academic researcher from Indiana University. The author has contributed to research in topics: Speech enhancement & Intelligibility (communication). The author has an hindex of 11, co-authored 36 publications receiving 945 citations. Previous affiliations of Donald S. Williamson include Ohio State University & Drexel University.
Papers
More filters
Journal ArticleDOI
Complex ratio masking for monaural speech separation
TL;DR: The proposed approach improves over other methods when evaluated with several objective metrics, including the perceptual evaluation of speech quality (PESQ), and a listening test where subjects prefer the proposed approach with at least a 69% rate.
Journal ArticleDOI
Time-Frequency Masking in the Complex Domain for Speech Dereverberation and Denoising
TL;DR: This paper performs dereverberation and denoising using supervised learning with a deep neural network and defines the complex ideal ratio mask so that direct speech results after the mask is applied to reverberant and noisy speech.
Proceedings ArticleDOI
Complex ratio masking for joint enhancement of magnitude and phase
TL;DR: This paper defines the complex ideal ratio mask (cIRM) that jointly enhances the magnitude and phase of noisy speech and employs a single deep neural network to estimate both the real and imaginary components of the cIRM.
Proceedings ArticleDOI
Speech dereverberation and denoising using complex ratio masks
TL;DR: A deep neural network is used to estimate the real and imaginary components of the complex ideal ratio mask (cIRM), which results in clean and anechoic speech when applied to a reverberant-noisy mixture and shows that phase is important for dereverberation, and that complex ratio masking outperforms related methods.
Proceedings Article
Towards Quantifying the "Album Effect" in Artist Identification.
TL;DR: Understanding the primary aspects of post-production of commercial recordings can attempt to model its effect on the acoustic features used for classification, and improve future systems for automatic artist identification.