Making my local LLM voice assistant faster and more scalable with RAG

Re:Making my local LLM voice assistant faster and more scala... by elevatedastalt on Friday June 14, 2024 @10:46PM (#40687119)

Cringe conversation. Why can't AIs just do stuff that you ask them to do without pretending to be human?

[ Parent | on HN ]

Re:Making my local LLM voice assistant faster and more scala... by zx8080 on Friday June 14, 2024 @10:56PM (#40687166)

Then how is it different from Excel/Word and shell/python scripts?

[ Parent | on HN ]

Re:Making my local LLM voice assistant faster and more scala... by exe34 on Saturday June 15, 2024 @01:15AM (#40687669)

it's much slower, gets things wrong, and insists on things that ain't so.

[ Parent | on HN ]

Re:Making my local LLM voice assistant faster and more scala... by WalterSear on Saturday June 15, 2024 @11:35AM (#40690404)

If it isn't, we have bigger problems.

[ Parent | on HN ]

Re:Making my local LLM voice assistant faster and more scala... by viraptor on Friday June 14, 2024 @10:59PM (#40687176)

Llama3 is very keen to be nice. I kind of wonder if that's due to better results on the chatbot arena (probably not, just a conspiracy theory I like). But with enough context available, you can definitely tweak the response in many ways. Give an example or two, tell it to be an emotionally detached HAL, you'll get what you want.

[ Parent | on HN ]

Re:Making my local LLM voice assistant faster and more scala... by colechristensen on Saturday June 15, 2024 @03:34AM (#40688113)

Because they’re trained to.

I hate the introduction to the response. That’s not even trying to be human, i don’t know something more like a deranged patronizing butler.

[ Parent | on HN ]