I have now Whisper and Piper as services for some pipelines: Like Podcast2Text and vice versa.

But surely thats amateur stuff - what have you set up on your home-server and/or cloud that integrates with AI/ML "intelligence"?