Skip to content

ElevenLabs Releases Scribe v3 and Scribe v3 Live Speech-to-Text Models

The DecoderRelayed· Maximilian Schreiner·· 12 hours agoPicksElevenLabs Scribe v3 Live uses conversational context to stop voice agents from cutting off callers mid-sentence
Original

What happened

ElevenLabs has released Scribe v3 for prerecorded audio and Scribe v3 Live for low-latency real-time use; both are available through ElevenAPI and ElevenAgents. Artificial Analysis benchmarks rank Scribe v3 among the most accurate transcription models tested, while finding it slower and more expensive than some rivals. The article says Scribe v3 Live uses conversational context to help voice agents avoid interrupting callers who pause briefly.

Why it matters

The article compares Scribe v3’s benchmark accuracy with its speed and price tradeoffs, while explaining why recorded-file tests do not capture Scribe v3 Live’s real-time behavior.

Read next

More changes in today’s briefingGoogle DeepMind Introduces Gemini 4 Hydrogen for Fast Agentic Workflows
This story is in 11/13 AI DailyRead the whole issue