Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Speaker Separation on LRS2 synthetic (test)
Loading...
14.2
SDR
VoiceFormer (Ours A+V+T)
-0.36
3.42
7.2
10.98
Jan 2, 2025
SDR
STOI
PESQ
Updated 4d ago
Evaluation Results
Method
Method
Links
SDR
STOI
PESQ
VoiceFormer (Ours A+V+T)
Video conditioning=tru...
2025.01
14.2
91.7
2.41
VoiceFormer (Ours A+V)
Video conditioning=tru...
2025.01
14.1
91.3
2.36
Visual Voice
Video conditioning=tru...
2025.01
10.8
88.4
2.16
Conversation
Video conditioning=fal...
2025.01
9.25
84
1.91
AVObjects
Video conditioning=tru...
2025.01
8.86
83.9
1.94
Noisy input
Video conditioning=fal...
2025.01
0.2
66.6
1.17
Denoiser
Video conditioning=fal...
2025.01
0.2
66.6
1.17
Feedback
Search any
task
Search any
task