You pay $1 once, render one minute of your own footage through the full pipeline, and see the 3 variants. If you subscribe, that $1 is credited to your first plan. One trial per person and per card.

Your plan gives you a monthly balance of source-video minutes. A 2-minute upload consumes 2 minutes of balance and still produces 3 variants. Unused failed renders are refunded automatically.

Moroccan Darija, Fusha Arabic and English — including transcription, subtitle timing and text rendering. Mixed-language videos work too.

Every word is aligned against the actual audio waveform (forced alignment), not just estimated by the AI. Timing errors of even half a second get corrected automatically.

Typically 10–20 minutes for a 1-minute video, depending on load. You can close the page — rendering happens in the cloud and the results wait in your dashboard.

Yes. Interactive Review Mode pauses after every stage so you can inspect the result, fix cuts and timings yourself, or type a correction for the AI — then approve to continue.

Yes. Upload your overlay library and a backdrop once; the AI catalogs them and places the right asset over the right sentence in every future render.

The minutes you spent on it are refunded to your balance automatically — you never pay for a failed render.

No. Every plan, including the $1 trial, exports clean HD clips with no watermark.