Shipmas Day 2: Podcast Anything App (Gemini 3)

Why It Matters

This lowers the friction for repurposing written content into polished audio, enabling creators and businesses to produce multi-voice podcast episodes rapidly for distribution or accessibility. It illustrates how LLMs plus multispeaker TTS can automate content transformation and editing workflows at scale.

Summary

The developer built a web app that converts uploaded documents (PDFs, markdown, text) into multi-voice podcast episodes by using Gemini 3 to generate scripts and a multispeech TTS API to produce audio. The interface offers controls for tone (roast, steelman, explain like a fifth grader), length ranges, and voice selection, and displays a timeline with download/playback options. In demos the tool turned a credit-card statement into a two-voice comedic roast and summarized a robotics benchmark paper as a short, kid-friendly explainer. The project was scaffolded with a Python TTS backend and a Next.js-style frontend, using Gemini API docs and an agent-driven planning step to assemble features quickly.

Original Description

Shipmas Day 2: Podcast Anything App (Gemini 3)

👊 Become a YouTube Member to Support Me:

https://www.youtube.com/c/AllAboutAI/join

My AI Video Course:

https://www.theaivideocourse.com/

🔥Open GH:

https://github.com/AllAboutAI-YT/

Business Inquiries:

[e-mail]