How can I use this to train char-embeddings on a Hindi-Language Corpus?

[skmalviya]

For example a Hindi corpus like -- indicnlp_corpus.

[hassyGo]

Hi @skmalviya ,

Thank you so much for your interests and I am sorry for my missing your question here.
Indeed, this is a crucial question.
At that time, I was using C++ in a naive way, and it is not trivial how to handle multi-byte characters.
I think now it is better to use Python to implement this for the language coverage.