Work

Exploring Generative Error Correction for Dysarthric Speech Recognition

Year: 2025

Type: conference-paper

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

Source: Interspeech (USB)

DOI: 10.21437/interspeech.2025-1553

Authors: Moreno La Quatra, Alkis Koudounas, Valerio Mario Salerno, Sabato Marco Siniscalchi

Language: English

Topic: Speech Recognition and Synthesis

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDGs: Reduced inequalities, Good health and well-being

Open Access status: green

Funders: European Commission, NextGenerationEU

Awards: G231133531, G8789261666, G972786147