question about manipulation types

[tuffy-studio]

Hello! I get question about manipulation types, according to Figure 3 in the paper:

it should be 12 types of manipulation in the dataset:
1.real
2.rtvc
3.FVRA-faceswap
4.FVRA-fsgan
5.FVRA-wav2lip
6.FVRA-fsgan-wav2lip
7.FVRA-faceswap-wav2lip
8.FVFA-faceswap
9.FVFA-fsgan
10.FVFA-wav2lip
11.FVFA-fsgan-wav2lip
12.FVFA-faceswap-wav2lip

But when I look through your csv file, there are only 8 types of manipulation(1, 2, 3, 4, 5, 10, 11, 12), where are the type 6,7,8,9?

[tuffy-studio]

And it seems that the RVFA samples are created using rtvc, which takes both the text and speech from the source video as input, rather than using the text from the source video and speech from another video in figure 3.