Models/ElevenLabs/Scribe v2 Realtime

Scribe v2 Realtime

ElevenLabs 📝 Transcription

Ultra-low latency (<150ms) live speech recognition. 93.5% accuracy across 90+ languages. WebSocket streaming with VAD.

Specifications

ModalitiesInput: audio  ·  Output: text
Speed●●●●● Blazing
FeaturesTools: No Streaming: No

Pricing

$0.0002 / generation

Platform credit pricing — live rates on the Models page.

Use Scribe v2 Realtime on Zubnet →

More from ElevenLabs

S
Scribe v1
Batch speech recognition with word-level timestamps and language detection across 99 languages.
📝 Transcription
M
Multilingual v2
Most advanced emotionally-aware speech synthesis with rich expression across 29 languages
🔊 Text-to-Speech
T
Turbo v2.5
Low-latency speech generation in 32 languages, optimized for real-time conversational AI
🔊 Text-to-Speech
F
Flash v2.5
Ultra-low latency TTS for real-time and conversational AI. ~75ms latency.
🔊 Text-to-Speech