Script for easily automating transcribing audio files on a remote GPU server with Whisper.
Specifically, sends an audio file to remote server, transcribes it, brings the resulting textfile back, and opens it.
mkdir -p ~/whisper
cd ~/whisper
mkdir inputs
mkdir outputs
python -m venv .venv
source .venv/bin/activate
pip install -U openai-whisper
cp config-default.yaml config.yaml
vi config.yaml
uv run main.py PATH_TO_AUDIO_FILE.m4a