Is it supervised, semi-supervised, or unsupervised?
Learn how the monolingual data should be and extract the monolingual corpus from the internet (find all the sources and how to crawl it -> might ask TA about it)
Try to see if it can be incorporated within low resource machine translation repo