Using Whisper for a fully local smart assistant implementation

[Nemesis533]

A while back I started working on a smart assistant that could be used fully locally - i tried a number of STT systems, but most were lacking one way or the other, especially performance wise; they were either great with a single language, or very fast, but none actually got decent results with multiple languages while also taking a reasonable time on reasonable HW (an old 2080ti or RTX 4060ti).

Whisper does example what I need it to and the turbo model has pretty good multilingual support (for English only, smaller models are great too).

this is the project I'm using it in and couldn't have gotten to this working state without whisper!

https://github.com/Nemesis533/Local_LLHAMA