Toward Equitable Machine Translation for Atypical Speech: An LLM Post-Correction Approach

A study presented at the 17th Conference of the Association for Machine Translation in the Americas (AMTA 2026) finds that large language models with 70B parameters or more consistently improve downstream machine translation quality for atypical speech, while smaller models lack the capacity to do so reliably. The researchers evaluated cascaded speech-to-text translation under dysarthric speech and simulated rhotacism across English-to-Spanish, English-to-Ukrainian, and English-to-Khmer directions, proposing a training-free LLM-based post-correction method as an accessible intervention.

Abstract Individuals with speech disabilities rely on everyday technologies powered by Automatic Speech Recognition ASR systems, yet these systems consistently fail them–producing significantly higher error rates that undermine the usefulness of voice assistants, hands-free devices, and machine translation pipelines. We conduct a multi-stage evaluation of speech impairment effects in cascaded speech-to-text translation, examining two distinct conditions: real dysarthric speech and simulated rhotacism. For dysarthria, we quantify ASR error rates. For rhotacism, we quantify error rates from a minimal-pair text substitution. We then analyze how impairment-induced errors propagate through downstream machine translation across three language directions English to Spanish, Ukrainian, Khmer , and propose a training-free LLM-based post-correction methodology as an accessible intervention. We find that larger LLMs 70B parameters or more consistently improve downstream translation quality even when surface-level corrections made by those LLMs remain modest, while smaller models lack the capacity to do so reliably. These results reveal a promising but scale-dependent path toward more equitable speech technology for users with atypical speech.- Anthology ID: - 2026.amta-research.3 - Volume: Proceedings of the 17th Conference of the Association for Machine Translation in the Americas Volume 1: Research Track /volumes/2026.amta-research/ - Month: - August - Year: - 2026 - Address: - Québec City, Canada - Editors: Eleftheria Briakou /people/eleftheria-briakou/unverified/ , Jeremy Gwinnup /people/jeremy-gwinnup/ , Shivali Goel /people/shivali-goel/unverified/ - Venue: AMTA /venues/amta/ - SIG: - Publisher: - Association for Machine Translation in the Americas - Note: - Pages: - 28–41 - Language: - URL: https://aclanthology.org/2026.amta-research.3/ https://aclanthology.org/2026.amta-research.3/ - DOI: - Cite ACL : - Grace Pasion, Ammon Shurtz, and Steve Richardson. 2026. Toward Equitable Machine Translation for Atypical Speech: An LLM Post-Correction Approach https://aclanthology.org/2026.amta-research.3/ . In Proceedings of the 17th Conference of the Association for Machine Translation in the Americas Volume 1: Research Track , pages 28–41, Québec City, Canada. Association for Machine Translation in the Americas. - Cite Informal : Toward Equitable Machine Translation for Atypical Speech: An LLM Post-Correction Approach https://aclanthology.org/2026.amta-research.3/ Pasion et al., AMTA 2026 - PDF: https://aclanthology.org/2026.amta-research.3.pdf https://aclanthology.org/2026.amta-research.3.pdf