Slowly decaying the fraction of aligned text interleaved with speech units during LLM fine-tuning improves speech-to-speech translation, most for low-resource languages.
Direct speech-to-speech translation with a sequence- to-sequence model,
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
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Scheduled Interleaved Speech-Text Training for Speech-to-Speech Translation with LLMs
Slowly decaying the fraction of aligned text interleaved with speech units during LLM fine-tuning improves speech-to-speech translation, most for low-resource languages.