Infinity Cast

Infinity Cast

Overview

Infinity Cast is a web application which hosts thousands of short-form AI-generated podcasts. Infinity Cast is a partially complete MVP passion and budget project.

Technologies Used

  • JavaScript React with TypeScript: frontend implementation
  • NodeJS: API implementation
  • Google Gemini: podcast topic and script generation
  • Google Gemini TTS: podcast audio generation
  • N8N: backend workers, jobs and admin APIs
  • Postgres: main data store for application
  • Amazon S3: podcast audio and cover art storage
  • Simple Diffuse: podcast cover art generation

Description

Infinity Cast aims to be a platform for both producing and consuming short-form podcasts. Short-form in the context of podcasts are podcasts ranging between 4 and 5 minutes in length.

Infinity Cast leverages the Google Cloud suite along with Gemini APIs to generate podcast topics, scripts and audio. The platform hopes to provide a unique experience for digesting content, primarily through unique playlist topics and endless topic generation. The eventual goal is to allow users to generate and share playlists with other users.

Key Features

  • Authentication and guest flows using JWT and OAuth
  • Creation of playlists
  • Podcast search with input text and filters
  • Generate podcast from description (currently disabled)

What I Learned

This was a fascinating project, and the first project I have heavily integrated with existing AI tools. I learned how to work effectively with generated code using Claude Sonnet integrated into my IDE to quickly get a full application up and running within weeks. I also explored other concepts and technologies that I had not used before including: Google Gemini APIs, OAuth integration, N8N workflows and Simple Diffuse APIs for image generation.

← Back to Projects