Aludel

Hey IH,

I'm Matteo, a UX designer based in Italy. Aludel is the first real app I've shipped.

It started from a pile of articles and documents I kept saving to read later and never actually reading. What I actually wanted was to listen to that stuff while walking, running, or driving, not sitting at a screen with more open tabs. So I started looking for a way to turn text into something I could put in my ears instead.

The privacy side mattered just as much. Some of what I work with is confidential, drafts, internal training material, things I didn't want to paste into someone else's cloud with no visibility into what happens next.

So Aludel turns text (or a URL) into a two-voice podcast conversation, with a few things I cared about getting right:

• Output language is decoupled from input, so a German source can become an Italian episode

• The generated script is editable before it becomes audio

• Both cloud providers (OpenAI, Anthropic, ElevenLabs) and fully local ones (Ollama, Piper) are supported, so it can run entirely offline if needed

• Built native for Apple Silicon with Tauri and Rust

One honest thing: I'm a designer, not a developer. I built most of Aludel through vibe coding with Claude rather than writing every line myself. I think that's part of the story, not something to hide. Someone who spends their days thinking about usability finally had a way to build the tool they kept wishing existed.

Aludel is in open beta right now with free Pro keys, live at aludel.matteobalocco.it. I'm looking for people who actually generate content in more than one language, or who care about keeping things local, to try it and tell me honestly where it breaks or where the UX doesn't make sense yet.

Happy to answer anything about the stack, the provider architecture, or the vibe-coding process.