ElevenLabs Releases Scribe v3 Speech Recognition Models

ElevenLabs has launched Scribe v3 and Scribe v3 Live, speech recognition models for pre-recorded and real-time audio, tuned for messy conditions such as crosstalk and background noise. The live version uses conversational context to avoid cutting callers off during brief pauses. Artificial Analysis benchmarks place Scribe v3 among the most accurate transcription models, though it is slower and pricier than rivals from Microsoft and xAI.