HyperCatcher
Surfaces content from your favorite podcasts
Description:
Today I release HyperCatcher Studio, a backend compliment to the HyperCatcher app. HyperCatcher Studio gives podcasters an easy way to add timely and actionable content directly into their podcast.
Using HyperCatcher Studio podcasters can add a way for users to subscribe to a newsletter, send a donation or see online content they reference in the podcast directly in their podcast app.
HyperCatcher Studio is the perfect way for podcasters to take control of their podcast and create the best experience for their listeners.
Background:
HyperCatcher Studio is the serious web app I've written. I wrote it using Flask and integrated heavily with Firebase. Using Firebase I was able to share account management and databases with the HyperCatcher podcast app which allowed me to write the whole web app in only a couple of weeks.
I'd envisioned building out something like HyperCatcher Studio before I started on the iOS Podcast App earlier this year, but I didn't know if I'd actually get as far as actually building it.
This is an exciting step towards the ultimate vision I have for a better podcasting experience, but there's still a long way to go.
I received a beta invite to OpenAi's GPT-3 Machine Learning Algorithm / API. I was able to feed podcast transcripts from Google Cloud's Speech to Text API into the GPT-3 a chunk at a time and get really good podcast summaries from this. From there I'm able to associate the summaries with podcasts and even segments of the podcast so that my users can jump to interesting parts of the podcast based on the summaries.
I'm still working on expanding this capability to more podcasts and on automating more of the process. You can read more about the process here: https://hypercatcher.com/2020/08/how-one-podcast-app-is-using-gpt-3-to-improve-your-podcast-listening-experience/
Comment