Switched to Fedora a year ago. And because I needed to transcribe interviews offline from time to time, I was looking for something like MacWhisper for Linux. When I couldn't find one, I decided to build one. Because in this day and age, why the hell not! And yes, AI coding-agents were used.

I work as a .NET Software Engineer and I despise Electron Apps, so I decided to go with a semi-native Stack, consisting of C# and Avalonia.Net (hardware accelerated trough SkiaSharp) for the frontend and C++ for low-level model bindings (CTranslate2 for translation, Nemotron 3 NeMo-Speech.cpp for Speaker Identification via JSON-RPC worker).

A friend liked the app and told me to go Freemium, so here we are. Works also on Windows btw.