With modern CPU’s supposedly shipping with ‘AI cores’: How long do you think it will take for a proper opensource, privacy respecting productivity tools(Something like whatever M$ copilot is supposed to be?) to be available?

Personally, i would love to see something like ‘Passive’ OCR integrated with the display server: the ability to pause any video and just select whatever text(even handwritten) there is naturally like it was a text document without any additional hassle will be really useful
Also useful in circumventing any blocks certain websites put on articles to prevent text from being copied

Or an AI grammar checker running natively for LibreOffice.

What are some AI tools you think should be developed for desktop Linux?

  • rufus@discuss.tchncs.de
    link
    fedilink
    arrow-up
    2
    arrow-down
    1
    ·
    edit-2
    9 months ago

    https://github.com/ggerganov/llama.cpp/ and several other free software projects already provide LLM interence on your laptop. I’m not sure which ones support AI powered OCR, I think that isn’t quite there yet, maybe stick with the traditional OCR solutions for a bit longer. But llama.cpp currently already supports 6 different multimodals models. And integrations to some IDEs also already exist, so something like Copilot is definitely there: https://continue.dev/

    However, I think the desktop side, and integration into other parts of the system are still entirely missing. I’m not aware of anyone working on it. I’d like speech regognition and nice TTS on the desktop.