In this project, you are going to build an LLM app that summarizes a podcast episode, identifies podcast guests, identifies key highlights, and more!
We are sure that there are many among you who love listening to podcasts and use them as a way to gain insights into different industries and technologies, as well as to learn from the lived experience of people all over the world. But there is limited time, and we end up listening to only some of them!
Many popular podcasts release 1–2 episodes a week, and it's hard to identify an episode that would appeal to each person. While many episodes provide show notes, additional links, and timestamps, they are not always helpful in understanding the unique aspects of the episode and providing a strong reason for each person to want to listen to it!
To combat this challenge, we are going to build a product that generates a newsletter every week that summarizes favorite podcasts for each person so that they can then choose which ones to listen to!
-
Part 1: Use a large language model from OpenAI to build the information extraction functionality, paired with a speech-to-text model for transcribing the podcast.
-
Part 2: Use a simple cloud deployment provider to easily convert the information extraction function to run on demand – This would be the app backend.
-
Part 3: Use ChatGPT from OpenAI as your coding assistant to create and deploy a front-end that allows users to experience the end-to-end functionality.