samhardyhey/blog-os-asr

OS ASR

Notebooks and scripts for the transcription of audio using open-source tools and models. See the accompanying blog posts here and here.

Install

• Conda env creation, python dependencies, low-level audio tools via create_env.sh
• Note some of the models referenced within 2_os_asr.ipynb and asr.py can be quite large

Usage

• Retrieve the first page's worth of podcast audio/transcripts via python retrieve_transcripts.py
• Transcribe audio files via python transcribe_audio_os.py <input_audio_file> <output_transcript_file.csv>