I look forward to more voice assistants that arent as laughably broken as Google/Alexa.
My Google Nest can’t even subtract from a timer (it can add, though!). Such a blatant surveillance device.
can’t even subtract from a timer (it can add, though!)
“Ok, Google, add -5 minutes to the timer”
It either explodes, or you start getting notifications in the past. Seems like a possible win, either way.
While this is certainly a cool concept, local voice assistants like this are currently a novelty. Cool to play around with, though!
You can expect around 5 seconds processing time to start generating the response to a basic question on a very basic model like Llama 3 8B.
For context, using Moondream2 (as recommended) on a RasPi 5, it takes around 50 seconds to process an image taken by the Camera and start generating a description.
Interesting, using whisper-fast on Home Assistant on my server computer takes like 2-3 seconds to process and delivery an output in English.
Useful in the smart home space.
Laughably broken in most other languages other than English, but then again, google and Alexa barely work in other languages.
Very slick. It looks like a thin wrapper around some pretty powerful tools, and I’m impressed that they’re still useful on such a low-power device.
I wrote an assistant a while back before Whisper was a thing, but now that I see what you’ve done, I’m going to have to go back and refactor.
Awesome job…I have a Pi5 arriving later this week so I now know what will be my first test project!
Are there any videos of Pi-card in action?
Why isn’t this an app?
The main reason for this is that I wanted to create a voice assistant that is completely offline and doesn’t require any internet connection. This is because I wanted to ensure that the user’s privacy is protected and that the user’s data is not being sent to any third party servers.
Weird reasoning, why constrain the assistant to pi’s power and memory when you can use it for interaction and utilize another, more powerful server at home
While the system is designed to be entirely local, it is also possible to easily connect it to some external APIs or services if you want to enhance the conversation
As an answer to your question I will remind you that unicycles still exist.
Ohh can I get it to use the voice from the Star Trek computer, Majel Barret?
Nice!
Now I need T.A.R.S
Why nobody is doing T.A.R.S? I would if I could but I can’t…