LipSync Generator
v1.0 — Frame-by-Frame Lip-Sync Generator
01 — Mouth shapes
Drop PNG files or click
Name each file by its sound:
ah.png oh.png mm.png rest.png
ah.png oh.png mm.png rest.png
Tip: Recommended names: rest ah oh ee oo mm pp ff ss th ch nn ww
02 — Audio track
Drop MP3 / WAV / OGG
The audio will be analyzed frame-by-frame
for phoneme detection at 24 fps
for phoneme detection at 24 fps
Allosaurus server (run server.py — works on any language, gibberish, singing)
check
Without server: heuristic analysis · With server: true IPA phonemes, any language
03 — Mouth shape mapping
Click any mouth to edit which sounds it represents
⚠ Mouth shapes appear here after uploading your PNG files in step 01
04 — Analyze & preview
FPS
Upload mouth PNGs and an audio file to begin
Phoneme timeline — 24 fps
05 — Export
Green screen
MP4
Solid #00FF00 background — use Ultra Key or chroma key in Premiere, Resolve, FCPX
Run analysis first
Output: .MP4 or .MOV files at source file resolution with audio baked in.