Scribe v2 — transcripts and captions in 90+ languages
Scribe v2 turns speech into text with speaker labels, word-level timings and sound events such as laughter and applause. Get an SRT and a transcript, or burn clean or bold captions straight into your clip.
ElevenLabs: expressive speech in 70+ languages, dialogue, music written to your cut, sound effects, dubbing and captions — the sound layer for every clip.
Open your galleryUse it through MCP
Captions and transcripts run on a clip you already have: open it in your gallery and pick captions or transcribe.
What it does
- Who said whatSpeakers are separated and labelled, with timings for every word.
- Captions burned inChoose a clean or bold style and get the clip back with captions, plus the SRT and the transcript.
- Your script, exact timingIf you already have the script, Mira aligns it to the audio instead of recognising it: your wording, timed to the frame.
- Sound eventsLaughter, applause, music and other events are marked in the transcript.
Specification
- Provider
- ElevenLabs
- Provider model
scribe_v2- Languages
- 90+ languages
- Limits on Mira
- source up to 30 min
- Output
- srt, json, mp4
- Where it works
- Gallery: an action on your clip or audio fileMCP tools:
transcribe_videoadd_captionsMira for Adobe panel (After Effects, Premiere Pro)
Cost in credits
2 credits per minute
Every started unit is billed in full.
| Example | Credits |
|---|---|
| Captions for a 1 min clip | 2 |
| Captions for a 10 min clip | 20 |
| Transcript of 10 min of audio | 10 |
The composer always shows the current price before you generate.
Questions about Scribe v2
Which languages does it recognise?
More than 90, and it detects the language by itself.
What files do I get?
An SRT caption file and transcript.json with speakers, words and timings. With burned captions, you also get the captioned clip.
Can it use my script?
Yes. Give it the script and Mira aligns it word by word, so the captions match your text exactly.
How much does it cost?
Captions cost 2 credits per minute of audio. A transcript without captions has its own rate; the table above shows both.
Can an AI agent caption my videos?
Yes: over the Mira MCP server, transcribe_video returns the transcript and SRT, and add_captions burns captions into the clip.