The Effectiveness of Time Stretching for Enhancing Dysarthric Speech for Improved Dysarthric Speech Recognition

Journal Article (2022)
Author(s)

Luke Prananta (Student TU Delft)

Bence M. Halpern (Nederlands Kanker Instituut - Antoni van Leeuwenhoek ziekenhuis, TU Delft - Multimedia Computing, Universiteit van Amsterdam)

Siyuan Feng (TU Delft - Multimedia Computing)

Odette Scharenborg (TU Delft - Multimedia Computing)

Multimedia Computing
Copyright
© 2022 Luke Prananta, B.M. Halpern, S. Feng, O.E. Scharenborg
DOI related publication
https://doi.org/10.21437/Interspeech.2022-190
More Info
expand_more
Publication Year
2022
Language
English
Copyright
© 2022 Luke Prananta, B.M. Halpern, S. Feng, O.E. Scharenborg
Multimedia Computing
Volume number
2022-September
Pages (from-to)
36-40
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

In this paper, we investigate several existing and a new state-of-the-art generative adversarial network-based (GAN) voice conversion method for enhancing dysarthric speech for improved dysarthric speech recognition. We compare key components of existing methods as part of a rigorous ablation study to find the most effective solution to improve dysarthric speech recognition. We find that straightforward signal processing methods such as stationary noise removal and vocoder-based time stretching lead to dysarthric speech recognition results comparable to those obtained when using state-of-the-art GAN-based voice conversion methods as measured using a phoneme recognition task. Additionally, our proposed solution of a combination of MaskCycleGAN-VC and time stretching is able to improve the phoneme recognition results for certain dysarthric speakers compared to our time stretched baseline.

Files

Prananta22_interspeech.pdf
(pdf | 4.33 Mb)
- Embargo expired in 01-07-2023
License info not available