Fairness in Dysarthric Speech Synthesis: Understanding Intrinsic Bias in Dysarthric Speech Cloning using F5-TTS

TL;DR

This paper examines biases in neural dysarthric speech synthesis, revealing that F5-TTS favors intelligibility over speaker similarity and prosody, highlighting the need for fairness-aware models to improve inclusivity.

Contribution

It provides an analysis of intrinsic biases in F5-TTS for dysarthric speech cloning, using fairness metrics to identify disparities across severity levels.

Findings

F5-TTS favors intelligibility over other speech qualities.

Biases vary across dysarthric severity levels.

Insights support development of fairness-aware speech synthesis.

Abstract

Dysarthric speech poses significant challenges in developing assistive technologies, primarily due to the limited availability of data. Recent advances in neural speech synthesis, especially zero-shot voice cloning, facilitate synthetic speech generation for data augmentation; however, they may introduce biases towards dysarthric speech. In this paper, we investigate the effectiveness of state-of-the-art F5-TTS in cloning dysarthric speech using TORGO dataset, focusing on intelligibility, speaker similarity, and prosody preservation. We also analyze potential biases using fairness metrics like Disparate Impact and Parity Difference to assess disparities across dysarthric severity levels. Results show that F5-TTS exhibits a strong bias toward speech intelligibility over speaker and prosody preservation in dysarthric speech synthesis. Insights from this study can help integrate fairness-aware dysarthric speech synthesis, fostering the advancement of more inclusive speech technologies.