Video caption generator

How captioning a video works here

The audio is extracted from your video and transcribed with word-level timing, then grouped into caption blocks. You get an editable transcript rather than a finished video, which keeps the captions separate from the picture.

Keeping them separate is usually what you want. A separate file can be corrected without re-encoding, replaced with a translated version, switched off by the viewer, and read by the platform — YouTube indexes uploaded caption text, so a captioned video becomes findable by what was said in it.

Video formats

mp4, mov, mkv, webm and avi all work, along with most other common containers. Only the audio track is used, so resolution and codec make no difference to the result or to the price.

If your video is very large, exporting just the audio first and uploading that instead will upload considerably faster and produce an identical caption file.

Uploading captions to a platform

• YouTube: Studio → Subtitles → Add language → Upload file → select your .srt.
• Vimeo: video settings → Distribution → Subtitles → upload the .srt or .vtt.
• A website using HTML5 video: reference the .vtt file from a <track> element.
• A video editor: import the .srt to burn captions into the picture if you need open captions.

Captions in other languages

The same video can produce caption tracks in any of the 76 supported languages. Because translation runs on the transcript, every language version shares the original timings, so you can upload several subtitle tracks against one video and they will all line up.