Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Identity-Preserving Talking Head Generation on CelebV-HQ hard (cross-video)
Loading...
47.7
Speaker Similarity
ID-LoRA (Ours)
33.868
37.459
41.05
44.641
Mar 10, 2026
Speaker Similarity
Face Similarity
LSE-D
LSE-C
CLAP Score
WER
Updated 1mo ago
Evaluation Results
Method
Method
Links
Speaker Similarity
Face Similarity
LSE-D
LSE-C
CLAP Score
WER
ID-LoRA (Ours)
Video Backbone=LTX-2
2026.03
47.7
87.4
8.49
3.9
0.363
11.3
CosyVoice 3.0 + WAN2.2
Video Backbone=WAN2.2
2026.03
39.1
89
11.4
1.5
0.249
36.2
Kling 2.6 Pro
Commercial Model=true
2026.03
38.5
85.4
9.49
3.47
0.316
12.1
ElevenLabs + WAN2.2
Video Backbone=WAN2.2
2026.03
35.7
89.4
11.86
1.72
0.238
15.4
VoiceCraft + WAN2.2
Video Backbone=WAN2.2
2026.03
34.4
89.2
10.6
1.33
0.258
42.7
Feedback
Search any
task
Search any
task