sync. labs brings clip, translate and dialogue edits into ChatGPT

sync. labs made its video editing flow available inside ChatGPT. Drop in any video and you can clip it, translate it or replace the dialogue, lipsynced in the original voice, without leaving the chat. To change something, describe it: swap a word, rewrite a line or add a language, and sync-3 matches the mouth to it.

To set it up, connect the official sync. labs plugin in ChatGPT and sign in to your Sync account through OAuth. No API key is needed. Ask ChatGPT to "Open Sync" and an embedded Studio opens, where you can browse your projects, upload or pick media, choose a model, compare the result with the original and download it.

There are four workflows. Lipsync video takes a video plus new audio, or a script and a voice. Image to video makes a talking image. Translate & Dub lets you search for the source and target language, with regional dialects such as Spanish for Argentina, Chile, Spain and Mexico. Edit Dialogue changes specific words in the original speaker's voice. It uses the surrounding sentence, matches room tone, takes pronunciation hints and allows up to 100 edits per preview.

ChatGPT shows the estimated cost in credits or dollars before you submit. New lipsync requests default to sync-3, and what you generate lives in your sync. labs account and shows up in Studio history.

The ChatGPT plugin first appeared in July 2026, so this is the editing flow arriving in chat, not a first launch. The workflows you get depend on your Sync plan and workspace, and generations use your Sync credits. Free accounts get up to 3 video generations a month, 15 seconds each.