I'm thinking of experimenting with my workflow to be more voice-oriented.

Specifically, I'll be looking to use Apple Voice Memos for speech / meeting recordings and then using some kind of voice-to-text tool to turn them into transcriptions (with some extra smarts around the edges that I could figure out if it works well enough).

Has anyone had any success with voice-to-text tools and algorithms? (Not necessarily Apple-specific)

I've previously looked into Mozilla Common Voice before but it wasn't really strong enough. I've heard there are tools that can detect multiple speakers these days - can anyone vouch for them? Is real-time transcription good enough yet to be a realistic option?