Automatic Speech Recognition on TORGO (Moderate)
3.27WERTTDS
Evaluation results
| Method | Links | |
|---|---|---|
| TTDS2026.02 | 3.27 | |
| ProtoDisent-TTSBackbone=Whisper-Medium, Augmentation=100% synthetic speech2026.02 | 4.19 | |
| SD-CTL2026.02 | 22.84 | |
| FMLLR-DNN2026.02 | 35.93 | |
| FS2D2026.02 | 36.8 | |
| Sys3Training=MFCC, Framework=DNN-HMM2026.06 | 65.69 | |
| Wav2Vec2Framework=SSL Models (Zero-shot Decoding)2026.06 | 68.03 | |
| Data2VecFramework=SSL Models (Zero-shot Decoding)2026.06 | 71.26 | |
| HuBERTFramework=SSL Models (Zero-shot Decoding)2026.06 | 72.88 | |
| Sys1Training=MFCC, Framework=DNN-HMM2026.06 | 95.89 | |
| Sys0Training=MFCC, Framework=DNN-HMM2026.06 | 115.41 |