Captions & AI · Auto captions

Auto captions, timed to every word

Turn speech into Persian or English captions with every word timed. Only a small audio track leaves your device.

What it does for you

Each word has its own start and end, which powers karaoke, highlight and pop animations.

The speech model is chosen for Persian, and sentences are rebuilt from punctuation and pauses.

Your browser turns the soundtrack into a 32 kbps Opus file; the video stays on your device.

Fixes misheard names and words, spelling, ZWNJ and punctuation while keeping each word’s timing.

Your browser

video.mp4 — stays here

capshak servers

Ready templates with fonts and motion

Translate and keep the timing

Spelling, half-spaces and names fixed

Describe the edit, it does the work

Edit and export for free, and pick a plan when you want AI captions. Nothing to install.