fameoai revised this gist . Go to revision
1 file changed, 29 insertions
fameo-notes.md(file created)
| @@ -0,0 +1,29 @@ | |||
| 1 | + | # Lip-sync quality checklist — a practical guide | |
| 2 | + | ||
| 3 | + | The goal here is a short clip that looks natural, finished in minutes rather than an afternoon. | |
| 4 | + | ||
| 5 | + | A one-page check I run before I share a talking video from [Fameo](https://fameo.ai). | |
| 6 | + | ||
| 7 | + | ## Image | |
| 8 | + | ||
| 9 | + | - [ ] Front-facing, with the mouth visible | |
| 10 | + | - [ ] Face large in the frame | |
| 11 | + | - [ ] No hands, toys, or deep shadows covering the lips | |
| 12 | + | - [ ] I have the right to use this still | |
| 13 | + | ||
| 14 | + | ## Audio and script | |
| 15 | + | ||
| 16 | + | - [ ] First version is 10–20 seconds | |
| 17 | + | - [ ] Pauses between sentences | |
| 18 | + | - [ ] Text I would actually say, or audio I recorded / licensed | |
| 19 | + | - [ ] The tool page matches the job (pet, cartoon, drawing, singing, or [multi-talk video](https://fameo.ai/lip-sync-generators/ai-multi-talk)) | |
| 20 | + | ||
| 21 | + | ## Review | |
| 22 | + | ||
| 23 | + | - [ ] Watched with sound | |
| 24 | + | - [ ] Watched with sound off — the mouth still reads as speech | |
| 25 | + | - [ ] Exported in HD | |
| 26 | + | - [ ] Caption or context makes it clear this is a generated talking still | |
| 27 | + | - [ ] No unauthorized likeness or impersonation | |
| 28 | + | ||
| 29 | + | Treat the first export as a draft; small script edits are cheaper than a second photo shoot. | |
Newer
Older