This is a handwriting-to-text converter! Just follow the instructions on the page and you're good to go :)<p>Background: I've been tutoring on the side for a while and it's apparent that the whole process can be smoothed out, with the end goal being an AI tutor buddy with a stylus interface. This is a little step in that direction.<p>As for implementation details, I forked excalidraw (at <a href="https://github.com/excalidraw">https://github.com/excalidraw</a>), got a gcp free tier instance running, and scraped together a Google K8s Engine cluster serving with torchserve. Luckily there's a great deal on the public preview of c3 cpus at the moment. For the model, I'm using trocr-base-handwritten ( <a href="https://huggingface.co/microsoft/trocr-base-handwritten" rel="nofollow">https://huggingface.co/microsoft/trocr-base-handwritten</a> ).<p>Let me know if anyone has any ideas, suggestions, and/or tips!
This is a handwriting-to-text converter! Just follow the instructions on the page and you're good to go :)<p>Background: I've been tutoring on the side for a while and it's apparent that the whole process can be smoothed out, with the end goal being an AI tutor buddy with a stylus interface. This is a little step in that direction.<p>As for implementation details, I forked excalidraw (at <a href="https://github.com/excalidraw">https://github.com/excalidraw</a>), got a gcp free tier instance running, and scraped together a Google K8s Engine cluster serving with torchserve. Luckily there's a great deal on the public preview of c3 cpus at the moment. For the model, I'm using trocr-base-handwritten ( <a href="https://huggingface.co/microsoft/trocr-base-handwritten" rel="nofollow">https://huggingface.co/microsoft/trocr-base-handwritten</a> ).<p>Let me know if anyone has any ideas, suggestions, and/or tips!
Neat project! Cool to see ideas merge like this. Obviously it’s really slow on my phone, but the idea is cool. Btw the conversion isn’t great. I wrote hello and its response Was way off