ElevenLabs releases Scribe v2 Medical for clinical audio
ElevenLabs says the speech-to-text model reduces word error rate on clinical audio by 18% compared with base Scribe v2, improving recognition of medication names, anatomy and pathology.
TLDR
ElevenLabs announced Scribe v2 Medical, a speech-to-text model fine-tuned for clinical audio. The company claims an 18% reduction in word error rate on clinical audio compared with its base Scribe v2 model, with better recognition of medication names, anatomy and pathology.
Combined views
—
1 Source, first seen ago
