- [Akka 2.6](https://doc.akka.io/docs/akka/2.6/typed/guide/actors-motivation.html) - - "cast" is word play with the [actor model](https://en.wikipedia.org/wiki/Actor_model) - ...and the feeling of casting/"magic" that using a voice-driven interface can inspire - This is the basis for the [[Graph visuals]] - [OpenAI Whisper](https://openai.com/research/whisper) (audio transcription) - Capture on Android via [Easy Recorder](https://play.google.com/store/search?q=easy%20recorder&c=apps&hl=en&gl=US) - Sync via [Syncthing](https://syncthing.net/) - [JVM folder watcher API](https://docs.oracle.com/javase/tutorial/essential/io/notification.html) detects sync'd files - Flask hosts Whisper for the Akka cast to access; I've looked into Docker, faster-whisper and other stuff but have stuck with my original - [Rasa open source](https://rasa.com/docs/rasa/#rasa-open-source) (natural language processing, entity extraction) - Here's a good example of [[Idea Provenance]] where I spoke groggily ("the back litter box was dirty") - ![[Pasted image 20240528120816.png]] - Inspired by [[Maps of Content (MOCs)]] - [[Why my Tinker Cast is not open-source|Why is this not open source?]]