axa-group/nlp.js

Is a separate corpus must for every language?

web3-anon opened this issue · 1 comments

Is there any way to auto-translate the English corpus during training/execution, or is it a must (the only way for multiple languages to work) to have a separate corpus for each language?

@web3-anon

To clarify, are you looking for a way to "translate" them with NLP.js or will you use a service / tool etc to translate for you? (if this is the case - NLP, does not "auto-translate" for you, tho you could use a service / tool / paid api etc to translate for you then add the translated results to the corpus using your own function etc...)

Or are you just curious if different languages added to corpus will require separate corpus files?

Let me know!