Exploring Generative Error Correction for Dysarthric Speech Recognition

Moreno La Quatra, Alkis Koudounas, Valerio Mario Salerno, Sabato Marco Siniscalchi
5/26/2025
cs.CLeess.AS

No official code found. Best implementation: MorenoLaQuatra/GER4Dys (6 stars, last commit 2025-05-27).

Abstract

Despite the remarkable progress in end-to-end Automatic Speech Recognition (ASR) engines, accurately transcribing dysarthric speech remains a major challenge. In this work, we proposed a two-stage framework for the Speech Accessibility Project Challenge at INTERSPEECH 2025, which combines cutting-edge speech recognition models with LLM-based generative error correction (GER). We assess different configurations of model scales and training strategies, incorporating specific hypothesis selection to improve transcription accuracy. Experiments on the Speech Accessibility Project dataset demonstrate the strength of our approach on structured and spontaneous speech, while highlighting challenges in single-word recognition. Through comprehensive analysis, we provide insights into the complementary roles of acoustic and linguistic modeling in dysarthric speech recognition

View on arXivView PDF

Code Implementations(1)

MorenoLaQuatra/GER4DysCommunity100%

This repo contains the code and information for the paper "Exploring Generative Error Correction for Dysarthric Speech Recognition" at INTERSPEECH 2025

61May 26, 20251 years ago

Similar Papers

Bridging ASR and LLMs for Dysarthric Speech Recognition: Benchmarking Self-Supervised and Generative Approaches

87% match2025

Perceiver-Prompt: Flexible Speaker Adaptation in Whisper for Chinese Disordered Speech Recognition

86% match2024

Enhancing AAC Software for Dysarthric Speakers in e-Health Settings: An Evaluation Using TORGO

86% match2024

Large Language Model Based Generative Error Correction: A Challenge and Baselines for Speech Recognition, Speaker Tagging, and Emotion Recognition

86% match2024

Empowering Dysarthric Speech: Leveraging Advanced LLMs for Accurate Speech Correction and Multimodal Emotion Analysis

85% match2024

Towards Inclusive ASR: Investigating Voice Conversion for Dysarthric Speech Recognition in Low-Resource Languages

85% match1 repo42025

Cite this paper

@article{quatra2025exploring,
  title  = {Exploring Generative Error Correction for Dysarthric Speech Recognition},
  author = {Moreno La Quatra and Alkis Koudounas and Valerio Mario Salerno and Sabato Marco Siniscalchi},
  year   = {2025},
  eprint = {2505.20163},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2505.20163v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2505.20163.svg)](https://codeofpaper.com/paper/2505.20163)

Discussion