- Published On
94 % on spoken benchmarks, but does it actually hear, or just transcribe? I tested two audio decision models on voices, echoes and sounds a transcript can't capture.
Optimized for Dark Mode
The website is optimized for dark mode to enhance your user experience. Switch to dark mode to enjoy it.
Hi! 👋 I'm a Data Scientist coming from Technical Systems and diving into GenAI✨
Find me on
My biggest and most important work.
A shadcn registry of streaming-aware Streamdown renderers for structured LLM output: Recharts charts from recharts-json fences with table and CSV/XLSX/PNG export, plus BPMN 2.0 process diagrams with progressive rendering, pan, zoom, fullscreen, and SVG/BPMN export.
German finetuned text-to-speech models: Kartoffelbox for expressive voice cloning, the compact MiraToffel miraTTS model, and Kartoffel-Orpheus-3B for natural German speech.
Want to get in touch with me?