symbolic languages like Chinese, Korean and Japanese needs to to be update

[vsatyamesc]

symbolic languages like Chinese, Korean and Japanese needs to to be update because the old fonts are not used much anymore and there's some new character too

[Robban1980]

I'm interested in the Japanese and Chinese models. I have only done small scale training in the past (years ago) for Japanese to use locally and testing. Are there any good resources on how I can improve the models with additional fonts?
And for example, what to look at during the training?