Scribe v2 Realtime
ElevenLabs · audio → text
ElevenLabs' real-time speech-to-text model with low-latency streaming transcription across 90+ languages. Call it via the native WebSocket endpoint wss /v1/speech-to-text/realtime (official SDKs work by changing base_url); billed by forwarded audio duration. Built for live captions, voice agents and real-time meeting notes.
Input$0 /M
Output$0 /M