Reinforcement learning with word-error and meaning-preservation rewards adapts an LLM-based speech recognizer to disordered speech better than supervised fine-tuning in this study.
Large language models as a proxy for human evaluation in assessing the comprehensibility of disordered speech transcription,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
eess.AS 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Speech Recognition With LLMs Adapted to Disordered Speech Using Reinforcement Learning
Reinforcement learning with word-error and meaning-preservation rewards adapts an LLM-based speech recognizer to disordered speech better than supervised fine-tuning in this study.