Summarizes the latest podcast episode, identifies podcast guests, and identifies key highlights/takeaways.
It uses GPT-3.5 LLM for creating the summary, identifying podcast and key highlights Whisper is used for speech-to-text conversion in order to transcribe the podcasts from audio to text. Modal is used for deploying the information extraction function on-demand.
Click here for the full demo generating podcasts on the go.