Look at camera. Say the line.
Look at camera. Say the line.
The look
Sampled frames retain the fictional presenter with coherent mouth and head movements. The male appearance now matches the selected male voice. Check precise lip-sync timing in playback. Generation took several minutes; an earlier attempt exceeded a four-minute timeout. Gemini 3.8 Flash transcribed the intended script and reported clear, calm male speech. This audio assessment cost a reported $0.00164 and is model-assisted. The sample was generated on 5 September. Open workflow uses the current model ID after a provider-name migration; the original sampled graph is retained below.
A face, a voice, a line to camera. Keep the script under 30 seconds. Match the presenter to the voice you picked. The saved take happens to be a workshop welcome — type the line you'd actually say.
Make it yours
- Open the graph — save first if the canvas already has your work. Undo brings yours back.
- Change Presenter look, Spoken script, and Delivery. Muse paints the face, MiniMax Speech reads, LongCat animates at 480p.
- The words are right, the face stays visible, the mouth follows the speech. Check lip-sync in playback — we don't claim frame-accurate timing. Generation can take several minutes; an earlier attempt blew a four-minute timeout. A Gemini 3.8 Flash transcript of the saved run reported $0.00164 (model-assisted). The open workflow uses current model IDs after a provider-name migration.
- The reviewed run reported at least $0.28. Video is listed at $0.03 per audio second at 480p, plus portrait and speech. Some provider price fields were omitted. Prices and results vary.
Download workflow · Original sampled graph · Explore MCP tools · awesome-noodles