This repo contains the files that are created as a result of running OpenAI's Whisper project on a video's audio (Ben Holmes' interview with Dan Abramov about React Server components). Below are instructions on how to use Whisper.
- Install
pyenv
- Install
pyenv-virtualenv
- Install Python 3.10.9:
pyenv install 3.10.9
- Setup a virtual envirnment for this project:
pyenv virtualenv 3.10.9 whisper
- use this virutalenv for this folder:
cd
to this directory and runpyenv local whisper
you should now see(whisper)
before your command prompt
- Install ffmpeg
Operating System Installation Instructions macOS Install Homebrew then run: brew install ffmpeg
in a terminalLinux (Debian) Run sudo apt update && sudo apt install ffmpeg
in a terminalWindows Install Chocolatey then run: choco install ffmpeg
in a terminal - Install the whisper python library:
pip install git+https://github.com/openai/whisper.git
- run
whisper PATH/TO/audio.wav
ffmpeg -i <infile> -ac 2 -f wav <outfile>