A simple Jupyter Notebook that takes a .mp3 file, parses the audio from it, and summarizes it.
I made this because I had a task where I had to summarize 4 different videos that were >2hrs in length each. I didn't want to spend the time watching them all, so instead I spent 1.5 hrs writing this script and saved a lot of time as a result.
Shoutout to OpenAI for providing Whisper (converts audio to writing), Huggingface for the summarizing language model.