Automatic Speech Recognition on Ikema kana (test)
20.94CERxls-r-300m
Evaluation results
| Method | Links | ||
|---|---|---|---|
| xls-r-300mArchitecture=wav2vec2-xls-r-300m, Number of parameters=300M, Pre-training languages=128, Batch size=16, Epochs=50, Learning rate=0.00032026.03 | 20.94 | 62.56 | |
| mms-1bArchitecture=mms-1b, Number of parameters=1B, Pre-training languages=1,406, Batch size=4, Epochs=50, Learning rate=0.00032026.03 | 21.86 | 63.22 | |
| xls-r-1bArchitecture=wav2vec2-xls-r-1b, Number of parameters=1B, Pre-training languages=128, Batch size=4, Epochs=50, Learning rate=0.00032026.03 | 24.07 | 66.61 |