BriefsToolAvailable now1 min read

ElevenLabs adds Scribe v3 and Scribe v3 Live speech to text models

ElevenLabs launched Scribe v3 and Scribe v3 Live, featuring lower word error rates, improved speaker diarization, and context-aware turn detection across 102 languages.

ElevenLabs adds Scribe v3 and Scribe v3 Live speech to text models — image from ElevenLabs
Image: ElevenLabs

Summary

ElevenLabs launched Scribe v3 for recorded audio and Scribe v3 Live for realtime transcription. Scribe v3 makes 27 percent fewer word errors than Scribe v2 across 102 languages, while Scribe v3 Live makes 43 percent fewer errors than Scribe v2 Realtime and introduces speaker diarization for realtime use. Both models are available now in ElevenLabs tools, the app, and ElevenAPI with a model ID update.

What changed

  • Scribe v3 makes 27% fewer word errors than Scribe v2 across 102 languages and includes speaker labels.
  • Scribe v3 Live is a low latency variant that makes 43% fewer errors than Scribe v2 Realtime, adding speaker diarization and turn detection.
  • Users can control outputs with keyterm prompting for up to 1,000 terms and custom transcript formatting rules in plain English.
  • Existing API integrations can upgrade by changing the model ID to scribe_v3 or scribe_v3_live.

Availability

Available now

“Scribe v3 Live is available today in ElevenAgents . Both models are also available in ElevenAPI , where switching from Scribe v2 is a one-line change to the model ID.”

Founder takeaway

Developers using ElevenLabs transcription can upgrade existing integrations by changing the model ID to use the new Scribe v3 or Scribe v3 Live models.

Source

ElevenLabsIntroducing Scribe v3 and v3 Live, our most accurate speech to text models

Announced Checked

Researched and drafted with AI from the official sources above, then checked automatically before publishing. Spotted a mistake? Tell us.

SharePost on XLinkedIn

Related

More briefs

All briefs
  1. LangSmith adds coding agent reporting and outcome tracking — image from LangChain

    ToolBeta

    LangSmith adds coding agent reporting and outcome tracking

    LangChain updates LangSmith with new reporting features that link AI coding agent spend to GitHub pull requests and GitLab merge requests.

    LangChain1 min read

  2. Reson8 releases Rumble-2-Flash as an open 0.5B speech AI model — image from Hugging Face

    ToolAvailable now

    Reson8 releases Rumble-2-Flash as an open 0.5B speech AI model

    Reson8 released Rumble-2-Flash, an open 0.5B-parameter speech recognition AI model supporting over forty languages.

    Hugging Face1 min read

  3. NVIDIA releases RTX Spark compact desktops for local AI — image from NVIDIA

    ToolAvailable now

    NVIDIA releases RTX Spark compact desktops for local AI

    ASUS, Dell, and MSI launched new RTX Spark desktops to run local AI agents, generation, and model fine-tuning on Windows.

    NVIDIA1 min read

One email, every Monday

Briefs are published here through the day. Monday Heist is the email: one useful AI workflow, the files to use it, and the catch. The first issue is being prepared; subscribers get it first. Free, and you can leave in one click.

One email. Every Monday.