[rohan_]

do you think it's feasible to create a Gen AI version of Alexa completely locally? unsure what exactly you're trying to say.

[KetoManx64]

HomeAssistant already does

[wkat4242]

Hmm yeah but it needs 30 seconds to give an answer. And that's with the LLM running on a GPU with HBM2 memory.

[ben_w]

It's not "Gen"erative, but is still a Transformer, and is kinda OK: https://github.com/openai/whisper

(I'm not too impressed with the error rate, but OTOH Alexa and Siri mishear me fairly often as well, so Whisper may be "good enough").

[bttrpll]

yes. that was the original design actually. and it worked. latency was a bit much though.

[cantrecallmypwd]

There are tools to make your own.

[throwawaymaths]

its feasible. if you want to try yourself you can start with the april-asr package.