We had some threads about languages recently. To be honest, it’s one of my favorite topics and I thought it was cool to see so many people talking about it.
So, I want to point out Lemmygrad communities I see about language for anyone interested:
Linguistics:
https://lemmygrad.ml/c/leftlang - somewhat active
https://lemmygrad.ml/c/linguistics - not active
Specific language learning:
https://lemmygrad.ml/c/learnchinese
September’s Korean study thread on c/korea
Translation:
https://lemmygrad.ml/c/translation
I’m not sure to what degree other people here are interested in seeing more language study topics be discussed around here but personally I would be very interested in more activity around this topic, such as helping each other study, producing translations together (I know that’s a big project), studying topics like language revitalization, language acquisition, etc.
Is anyone else here enthusiastic about this kind of topic?
If you know other languages good enough like Spanish or Chinese or any other major language that is useful.
It’s hard to make the code good enough to accurately break down individual sentences on its own.
Yeah, that would be the problem with such a solution. It’s hard to analyze human language using a machine.
I was using Google/Baidu and argos ML models to translate the text in large sections and it can do it fairly accurately because it’s trained on large corpus’s of natural language and it can operate given context.
Trouble is for my application I need to use libraries like THULAC to separate the individual parts of speech in a sentence and then hopefully translate the nouns and verbs that need to be translated for the user automatically. But that is something the existing models are just not built for. (or maybe they can be retrofitted/adapted?) So I may need new models and combine that with some hand coded logic to get the best automatic result. Or the app/tool will just need to rely on more manual methods meaning you might need to do it yourself upfront.