Automatic Speech Recognition on TORGO (Severe)
22.15WERProtoDisent-TTS
Evaluation results
| Method | Links | |
|---|---|---|
| ProtoDisent-TTSBackbone=Whisper-Medium, Augmentation=100% synthetic speech2026.02 | 22.15 | |
| TTDS2026.02 | 23.3 | |
| FMLLR-DNN2026.02 | 43.29 | |
| FS2D2026.02 | 55.88 | |
| SD-CTL2026.02 | 68.24 |