Speech recognition for English medical speech in Indian clinics.
A 5B ASR model for short-form English medical audio as spoken in Indian healthcare settings: a Whisper large-v3 encoder joined to a MedGemma 4B decoder by a lean projector layer. Handles clips under 30 seconds.
Install
Accept the terms on Hugging Face, then download
hf download ekacare/parrotlet-a-en-5b Links
Topics
asrspeech-recognitionmedical-transcriptionwhispermedgemma