Shotcraft
Voice and speech

Scribe v2

Speech-to-text model covering 90+ languages with word- and character-level timestamps and speaker diarization for up to 32 speakers.

A speech model by ElevenLabs · read 2026-08-14 · sourcing high

Weights
Not released
API
Yes
Commercial use
Permitted subject to conditions

It adds entity detection and redaction, audio event tagging, keyterm biasing and multichannel handling. A separate realtime variant streams at roughly 150ms.

Commercial use — Permitted subject to conditions

Standard ElevenLabs terms; commercial use permitted on paid plans.

  • Depends on your plan tier. Standard ElevenLabs terms; commercial use permitted on paid plans

The conditions above were extracted from the licence notes recorded during research, and each one shows the sentence it came from so you can check the reading. Read the licence before you rely on any of this. Nothing on this page is legal advice, and this site does not tell you a model is cleared for commercial use.

Specification

Vendor figures, read 2026-08-14Read from ElevenLabs's own documentation on 2026-08-14. Vendors change these between model versions without announcing it, so check theirs before you commit a shot to a number. Sourcing for this entry is strong: the figures came from pages we retrieved and read. Check the vendor's current documentation

What it takes in
audio, video
What it makes
text
Controls
timestamps, streaming-low-latency
Weights
Not released
API
Yes
Pricing
Scribe v2 $0.22 per hour, plus $0.070/hour for entity detection and $0.050/hour for keyterm prompting. Scribe v2 Realtime $0.39 per hour.
Released or updated
not established
Languages
90, not individually listed by the vendor

Strengths

  • 90+ languages with diarization up to 32 speakers
  • Word and character-level timestamps, useful for caption and subtitle workflows
  • File uploads up to 5GB (2GB via source URL)
  • Cheap at $0.22/hour

Weaknesses

  • Extra per-hour charges for entity detection and keyterm prompting
  • Realtime variant costs ~1.8x the batch rate
  • No published word error rate on the pages fetched

Best for

Where the sources disagree

Recorded rather than resolved. Picking the tidier number would hide the fact that the vendor's own pages do not agree.

What to understand before using it

Explainers naming the mechanisms this model's own documentation mentions.

MCP servers that front ElevenLabs

Servers in the media MCP directory whose own description or homepage names this vendor. Absence of a server here is not evidence that none exists: most publish no tool list at all.

26 further servers name ElevenLabs with lower confidence and are not listed, because a server naming a vendor may be using it for text rather than media.

See it from the vendor

Shotcraft hosts no sample output. Judging a model means watching its own showreel with your own eyes, and a clip re-hosted here would be a clip we chose.

Nearby in the directory

ElevenLabs Agents

A hosted conversational voice stack that chains a fine-tuned speech-to-text model, a selectable LLM, low-latency TTS and a proprietary turn-taking mod

Voice and speech

ElevenLabs Dubbing (v2)

End-to-end dubbing that transcribes, translates and regenerates speech in the source speaker's voice across 90+ languages, with dialect variants in v2

Voice and speech

Eleven v3

ElevenLabs' most expressive text-to-speech model, aimed at emotionally varied narration rather than lowest latency. It supports natural multi-speaker

Voice and speech

Sources

Read on 2026-08-14. A source that has changed since then will not be reflected here until the next pass.