Speech Reconstruction on MLS non-English OOD (700 utterances (7 languages))
3.16UTMOSDTM-Codec
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| DTM-Codecbps=800, Frame Rate (Hz)=50Hz, Rate Control Strategy (FFR/VFR)=VFR2026.06 | 3.16 | 2.5 | 92 | 83 | |
| DTM-Codecbps=640, Frame Rate (Hz)=40Hz, Rate Control Strategy (FFR/VFR)=VFR2026.06 | 3.15 | 2.3 | 91 | 79 | |
| DTM-Codecbps=400, Frame Rate (Hz)=25Hz, Rate Control Strategy (FFR/VFR)=VFR2026.06 | 3.09 | 1.94 | 88 | 65 | |
| DTM-Codecbps=1280, Frame Rate (Hz)=80Hz, Rate Control Strategy (FFR/VFR)=FFR2026.06 | 3.07 | 2.72 | 94 | 87 | |
| DTM-Codecbps=640, Frame Rate (Hz)=40Hz, Rate Control Strategy (FFR/VFR)=FFR2026.06 | 3.05 | 2.19 | 90 | 72 | |
| DTM-Codecbps=800, Frame Rate (Hz)=50Hz, Rate Control Strategy (FFR/VFR)=FFR2026.06 | 3.04 | 2.31 | 91 | 75 | |
| DTM-Codecbps=1280, Frame Rate (Hz)=80Hz, Rate Control Strategy (FFR/VFR)=VFR2026.06 | 3 | 2.66 | 93 | 90 | |
| FlexiCodecbps=9052026.06 | 2.99 | 2.33 | 90 | 84 | |
| DTM-Codecbps=400, Frame Rate (Hz)=25Hz, Rate Control Strategy (FFR/VFR)=FFR2026.06 | 2.97 | 1.86 | 87 | 61 | |
| VARSTokbps=4022026.06 | 2.6 | 1.49 | 81 | 44 |