search

joonson / yousaidthat

You Said That?: Synthesising Talking Faces from Audio
69 stars 19 forks source link

Voxceleb2 header? #2

Open mrgloom opened 5 years ago

mrgloom commented 5 years ago

What does these fields mean in Voxceleb2 header?

    Offset    :     -2
    FV Conf   :     16.303  (1)
    ASD Conf  :     6.201
joonson commented 5 years ago

Offset : Audio-to-video offset in # frames (using SyncNet) FV Conf : Face Verification confidence (using VGGFace2) ASD Conf : Active Speaker Detection confidence (using SyncNet)