Phoneme Pronunciation Assessment on speechocean762 official (test)
0.679PCCGOPT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GOPTAcoustic Model=PAII-A2022.05 | 0.679 | 0.069 | |
| HIA2026.01 | 0.657 | 0.076 | |
| Gradformer2026.01 | 0.646 | 0.079 | |
| SpeechBlender-AugCategory=Our Proposed Augmentation Method, Augmentation Technique=SpeechBlender-Aug2022.11 | 0.63 | 0.085 | |
| HiPAMA2022.11 | 0.616 | 0.084 | |
| HiPAMA2026.01 | 0.616 | 0.084 | |
| GOPTAcoustic Model=Librispeech2022.05 | 0.612 | 0.085 | |
| Gong et al.2022.11 | 0.612 | 0.085 | |
| GOPT2026.01 | 0.612 | 0.085 | |
| Multi-task Learning (w. Transformer)Category=Baselines, Architecture=Transformer, Learning Mode=Multi-task Learning2022.11 | 0.61 | 0.085 | |
| Gong et al. (implemented baseline)implementation=re-implemented baseline2022.11 | 0.609 | 0.086 | |
| LSTMAcoustic Model=Librispeech2022.05 | 0.591 | 0.089 | |
| LSTM2026.01 | 0.591 | 0.089 | |
| Text-augmentationCategory=Augmentation Baselines, Augmentation Technique=Text-augmentation2022.11 | 0.59 | 0.092 | |
| GOP-augmentationCategory=Augmentation Baselines, Augmentation Technique=GOP-augmentation2022.11 | 0.59 | 0.09 | |
| LSTM2022.11 | 0.587 | 0.089 | |
| Speechocean762-TrainCategory=Baselines, Training Dataset=Speechocean762-Train2022.11 | 0.58 | 0.09 | |
| Multi-task Learning (w. LSTM)Category=Baselines, Architecture=LSTM, Learning Mode=Multi-task Learning2022.11 | 0.56 | 0.089 | |
| Human2026.01 | 0.555 | — | |
| UOR2026.01 | 0.52 | 0.12 | |
| SVRAcoustic Model=Librispeech2022.05 | 0.45 | 0.16 | |
| SVR2026.01 | 0.45 | 0.16 | |
| RFAcoustic Model=Librispeech2022.05 | 0.44 | 0.13 | |
| RF2026.01 | 0.44 | 0.13 |