Is there a way to use an LLM and (x,y) coordinates to improve swipe typing?

grill_gus

I am looking to build more advanced gesture typing which takes into account the previously typed words as well as the (x,y) coordinates of gestures, thus improving the swipe algorithm manyfold. Where do I start building this? Right now I do have two model approaches where, I get predictions for next word from the LLM. I construct a list of possible words from the gesture typing possible. I can then use both of them to find the best possible...

Originally asked by Vaibhav Maheshwari on Stack Exchange (ai) (CC BY-SA 4.0).

uphilltech

Good thread on this — worth a read before you pay anyone to build it.

Discussion: https://ai.stackexchange.com/questions/50224/is-there-a-way-to-use-an-llm-and-x-y-coordinates-to-improve-swipe-typing.

Want it set up for you? UphillTech.