Phone Level Pronunciation Scoring On
Métriques
Pearson correlation coefficient (PCC)
Résultats
Résultats de performance de divers modèles sur ce benchmark
Nom du modèle | Pearson correlation coefficient (PCC) | Paper Title | Repository |
|---|---|---|---|
| GOPT-Librispeech | 0.61 | Transformer-Based Multi-Aspect Multi-Granularity Non-Native English Speaker Pronunciation Assessment | - |
| SpeechBlender + LSTM | 0.63 | SpeechBlender: Speech Augmentation Framework for Mispronunciation Data Generation | - |
| HiPAMA-Librispeech | 0.62 | Hierarchical Pronunciation Assessment with Multi-Aspect Attention | - |
| GOPT-PAII | 0.68 | Transformer-Based Multi-Aspect Multi-Granularity Non-Native English Speaker Pronunciation Assessment | - |
| 3MH | 0.693 | A Hierarchical Context-aware Modeling Approach for Multi-aspect and Multi-granular Pronunciation Assessment | - |
| E2E-R | 0.68 | Fine-Tuning Self-Supervised Learning Models for End-to-End Pronunciation Scoring | |
| GOP | 0.45 | speechocean762: An Open-Source Non-native English Speech Corpus For Pronunciation Assessment | - |
0 of 7 row(s) selected.