ManWav: The First Manchu ASR Model
| Autor: | Seo, Jean, Kang, Minha, Byun, Sungjoo, Lee, Sangah |
|---|---|
| Rok vydání: | 2024 |
| Předmět: | |
| Druh dokumentu: | Working Paper |
| Popis: | This study addresses the widening gap in Automatic Speech Recognition (ASR) research between high resource and extremely low resource languages, with a particular focus on Manchu, a critically endangered language. Manchu exemplifies the challenges faced by marginalized linguistic communities in accessing state-of-the-art technologies. In a pioneering effort, we introduce the first-ever Manchu ASR model ManWav, leveraging Wav2Vec2-XLSR-53. The results of the first Manchu ASR is promising, especially when trained with our augmented data. Wav2Vec2-XLSR-53 fine-tuned with augmented data demonstrates a 0.02 drop in CER and 0.13 drop in WER compared to the same base model fine-tuned with original data. Comment: ACL2024/Field Matters |
| Databáze: | arXiv |
| Externí odkaz: |
|