Media teams and developers needing accurate transcripts in 90-plus languages at $0.22 an hour, with a few free hours to test.

Kenning's take

What it does

Scribe v2 is ElevenLabs' transcription model for recorded audio, announced on 9 January 2026. It improves on Scribe v1 with long recordings and adds keyterm prompting, entity detection, speaker labels and word-level timestamps. It is available through the ElevenLabs API and powers subtitles and transcripts in the ElevenLabs creative platform and Studio.

• Transcription in 90+ languages
• Keyterm prompting for brand and technical words
• Entity detection for sensitive data
• Speaker diarisation up to 32 speakers
• Word-level timestamps and audio event tags

Who it suits. Developers, media teams and creators who need accurate transcripts and subtitles at scale.

Key facts

Category

AI models

Part of

ElevenLabs (ElevenReader and ElevenLabs app)

Made by

ElevenLabs

Based in

United States

Released

9 January 2026

How to get it

through the ElevenLabs API and in ElevenLabs Studio and the creative platform, on free and paid plans

Works on

API, Web

In the UK

Yes

Free plan

4 hours 30 minutes of transcription on the Free plan

API price

$0.22 per hour of audio; keyterm prompting $0.05 per hour, transcript editing $0.07 per hour

FAQ

Facts from the maker's own pages and the sources listed. Checked 2026-10-11.