PDFs turned into a podcast automatically

What it solves

Reports, papers, manuals, internal documentation: material that piles up unread because it forces you to sit in front of a screen. This system turns it into audio, ready to listen while driving, training or doing something else. The output is delivered as a real podcast — with its own RSS feed — so it can be added to Apple Podcasts, Spotify or any similar app.

How it’s built

The process goes from form to RSS feed, with no manual steps:

1. Upload — the user uploads the PDF through a web form and the workflow kicks off.
2. Extraction and cleanup — the text is extracted, headers and page numbers are stripped, and it’s split into logical chunks sized to fit the voice API’s input limit.
3. Quota check — before spending any credit, the system queries the month’s usage on Google Cloud to stay within the free tier and avoid surprise charges.
4. Synthesis — each chunk is rendered into natural voice with Google Text-to-Speech and all fragments are concatenated into a single MP3.
5. Publishing — the audio is uploaded to Cloudflare R2 and an RSS feed is generated with the metadata Apple Podcasts and Spotify require.
6. Notification — the user gets an email with the feed link, the generated sections and the quota consumption percentage.

Why it’s interesting

What’s interesting isn’t the text-to-audio conversion — that’s commodity — but delivering it as a podcast that plugs into any app, without building an audio platform. The real cost is practically zero: Google’s free tier covers one million characters per month, and Cloudflare R2 doesn’t charge for egress traffic. For a company it’s a simple way to make internal documentation accessible, or to distribute training content without paying for a distribution platform.

The workflow, in the open

The full workflow is published on GitHub: you can download it, import it into your own n8n and see exactly how it’s built, node by node → pdf-a-podcast.json