Continual pre-training on Sinhala followed by Dhivehi fine-tuning achieves 12.89% WER, but KenLM decoding contributes 27+ absolute WER points, dwarfing the 0.61-point transfer gain.
Cross-language end-to-end speech recognition for low- resource tujia language,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
From Sinhala to Dhivehi: Cross-Lingual Transfer Learning for Low-Resource Speech Recognition
Continual pre-training on Sinhala followed by Dhivehi fine-tuning achieves 12.89% WER, but KenLM decoding contributes 27+ absolute WER points, dwarfing the 0.61-point transfer gain.