Automatic Speech Recognition on MuST-C En-Es v1.0 (test)
62.2ASR chrFUnitY + t-mBART
Evaluation Results
| Method | Links | |
|---|---|---|
| UnitY + t-mBARTDecoder layers=12L->2L, Pre-training=wav2vec2.0 (encoder), t-mBART2022.12 | 62.2 | |
| UnitYDecoder layers=6L->6L, Pre-training=wav2vec2.0 (encoder)2022.12 | 61.8 | |
| S2UT + u-mBARTPre-training=wav2vec2.0 (encoder), u-mBART2022.12 | 61.5 | |
| S2TT-TTSPre-training=wav2vec2.0 (encoder)2022.12 | 61.3 | |
| S2SpecT2 + t-mBARTDecoder layers=12L->6L, Pre-training=wav2vec2.0 (encoder), t-mBART2022.12 | 61.1 | |
| S2SpecT2Decoder layers=6L->6L, Pre-training=wav2vec2.0 (encoder)2022.12 | 60.6 | |
| ASR-MT-TTSPre-training=wav2vec2.0 (encoder)2022.12 | 58.7 |