How to programatically identify the pitch accent of a term (bringing pitch accent to emacs)

hi everybody.

I have been writing a module to bring yomichan/morphman style features to emacs ([https://github.com/dmgerman/yomikun](https://github.com/dmgerman/yomikun) \–open source)

So far I am able to do tokenization (including compounds), status tracking and dictionary. Though, it is still work in progress.

I want to add pitch accent support. But I have not been very successful at finding info on how to do it.

Does anybody know of a good source of information for how to do it?

1. Given a sentence, how the pitch accent of each term is determined
2. How/where to find the data necessary

I see there is an NHK pitch accent csv file floating around with pitch info. But I have not been able to find documentation about it.

thank you.