A hybrid HMM system using self-supervised features and extra text/audio data beats fine-tuned Whisper by 32 percent relative WER on Scottish Gaelic broadcast speech.
Hybrid models [25] can effectively combine acoustic models trained on transcribed data with language mod- els trained on all available text
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
-
A Practitioner's Guide to Building ASR Models for Low-Resource Languages: A Case Study on Scottish Gaelic
A hybrid HMM system using self-supervised features and extra text/audio data beats fine-tuned Whisper by 32 percent relative WER on Scottish Gaelic broadcast speech.