How to use it
Choose a video up to 90 seconds, select its spoken language and generate captions for 1 credit. Review the wording and timing, then export MP4 or SRT. Audio is sent for AI transcription.
Video & subtitles
Transcribe short videos, edit word timing, and export a captioned MP4 or SRT file.
Choose video → Generate captions → Edit text and timing → Download
Failed jobs show their credit refund status.
Whisper large-v3 · word timestamps
The original video stays in your browser. Only a lightweight extracted audio track is sent to Replicate's Whisper model for transcription, and the uploaded file is deleted immediately afterward.
Upload a video to preview caption style and timing here.
TOOL GUIDE
Content reviewed · 2026-10-12
Transcribe short videos, edit word timing, and export a captioned MP4 or SRT file.
Choose a video up to 90 seconds, select its spoken language and generate captions for 1 credit. Review the wording and timing, then export MP4 or SRT. Audio is sent for AI transcription.
Music, overlapping voices and names can cause errors. Review the whole video before publishing. Keep this tab open during MP4 export.
The main task runs in the browser, while a named model or AI service may be contacted for part of the feature.
A few tools that fit naturally with this task.
Submit a product URL or a plain-language request to compare sellers and similar products, estimate a fair price range, and surface purchase checks.
Automatically detect the input language and correct grammar, spelling, typos, punctuation, and awkward phrasing.
Transform photos of people, pets, objects, and places into character cards, posters, figures, and more.