Video-to-Speech Synthesis
Benchmarks
DatasetSOTA methodMetricTrendResultsLast updated
0MCD
13
Feb 26, 2026
79Sim-O
11
Feb 26, 2026
0.87Sim-O
11
Feb 26, 2026
3.94MOS-S
10
Feb 26, 2026
3.921UTMOS
7
Feb 26, 2026
4.031UTMOS
7
Feb 26, 2026
3.17MOS (Naturalness)
4
Apr 20, 2026
0.65GE2E
4
Feb 26, 2026
1.3PESQ
4
Feb 26, 2026