Inspiration
I was born in Jacaltenango, Guatemala where Maya popti is still spoken. My parents and family still speak our native language of popti, but with time the language is being spoken less and less. Preserving languages like popti is essential to preserve indigenous cultures.
What it does
Translates source language to a targeted language - In my attempt, it was Popti to Spanish.
How we built it
Using OpenNMT an open-source ecosystem for neural machine translation and neural sequence learning. I used their pipeline to train the model using self-made datasets.
Challenges we ran into
When training the model issues with cuda and torch were encountered that did not allow me to finish the training of the model.
Accomplishments that we're proud of
Utilizing an open-source product like OpenNMT and creating a dataset were small victories in the project.
What we learned
I learned about PyTorch and OpenNMT. Also working with Jupyter Notebook was a learning experience.
What's next for Popti Translator
I would like to finish the training process and create a UI/App that can be implemented.
Log in or sign up for Devpost to join the conversation.