Skip to content

Latest commit

 

History

History

MedASR Notebooks

  • Quick start with Hugging Face - This notebook demonstrates how to quickly get started with MedASR, Google's open Automatic Speech Recognition (ASR) model fine-tuned for the medical domain. It shows how to load the MedASR model from Hugging Face and use the transformers library pipeline to perform speech-to-text transcription on an example audio file.

  • Quick start with Model Garden - This notebook demonstrates how to use MedASR as a service deployed on Vertex AI. It shows how to use the service API to perform speech-to-text transcription in online or batch workflows.

  • Fine-tune with Hugging Face - This notebook shows how to fine-tune the model locally using Hugging Face libraries.