PDF to Podcast

PDF to Podcast transforms any PDF document into a podcast-ready audio episode using advanced AI text-to-speech (TTS) providers. Upload a PDF, select your preferred voice and provider, and receive an MP3 and a ready-to-use RSS feed for your podcast app.

🚀 Quick Launch

Deploy this app to your favorite platform with one click:

Docker:

docker run -d -p 3000:3000 -e GEMINI_API_KEY=your_key ghcr.io/knoksen/pdf-to-podcast:latest

Heroku:

Fly.io:

fly launch --image ghcr.io/knoksen/pdf-to-podcast:latest

📝 Features

• PDF Upload: Drag & drop or select a PDF, up to 50MB.
• Provider Selection: Choose TTS engine: gTTS (free), OpenAI, or Azure.
• Voice Selection: Customizable voice (where supported).
• Episode Management: See, play, and download all generated episodes.
• Podcast-Ready: Instant RSS feed for any podcast app.
• Modern UI: Built with React, TailwindCSS, and best practices.
• API-First: Easily extensible for new TTS providers.

🛠️ Architecture & Tech Stack

• Frontend: React (Vite), TypeScript, TailwindCSS
• Backend: Node.js (Express/Fastify), REST API
• TTS Providers: gTTS (Google Text-to-Speech), OpenAI, Azure Cognitive Services
• Storage: Local filesystem (default), configurable for cloud (e.g., S3)
• Deployment: Vercel, Netlify, Railway, Render, Docker, etc.
• CI/CD: GitHub Actions (recommended)
• RSS Generation: Automatic and standards-compliant

⚙️ Configuration & Environment

Create a .env.local file in the root with:

GEMINI_API_KEY=your_gemini_api_key
# Optionally:
# OPENAI_API_KEY=your_openai_key
# AZURE_API_KEY=your_azure_key
# AZURE_REGION=your_azure_region
# STORAGE_DIR=/path/to/episodes

• GEMINI_API_KEY is required for Gemini-based TTS.
• Add other provider keys as needed.

🧑‍💻 Local Development

Prerequisites:

• Node.js 18+
• npm

Steps:

git clone https://github.com/knoksen/pdf-to-podcast.git
cd pdf-to-podcast
cp .env.local.example .env.local # then edit the file
npm install
npm run dev

Visit http://localhost:3000

🗒️ Usage

1. Upload your PDF.
2. Choose a provider and voice (if available).
3. Click Generate Episode.
4. When processing is complete, play or download the MP3.
5. Use /rss.xml in your podcast app.

🧩 API (for Developers)

• POST /api/episodes – Upload a PDF and metadata
• GET /api/episodes – List all podcast episodes
• GET /api/episodes/:id/audio – Download episode audio
• GET /rss.xml – Get podcast RSS

Extending:
Add your own TTS provider by implementing a new provider in /services/tts/ and updating the provider options.

📦 Docker

Build your own image:

docker build -t pdf-to-podcast .
docker run -d -p 3000:3000 -e GEMINI_API_KEY=your_key pdf-to-podcast

🧪 Testing

npm run test

🤝 Contributing

PRs and issues welcome! See CONTRIBUTING.md for details.

📄 License

MIT

✨ Credits

• Vercel, Netlify, Railway, Render, Azure Cognitive Services, OpenAI, gTTS, Google Gemini

🗣️ Questions?

Open an issue or start a discussion.