hello,
I would like to have a chatbot that can communicate in different languages. Most of the documentation I have found uses google translator or similar things, but we all know that it is not the best and so I am looking for other options. I have also read that they should be trained separately and run independently. My question is: is there a way that the voice chatbot can be used in different languages but doesn’t spend a lot of time having to train each language separately (I don’t know if it’s possible). Maybe something related to a database or something like that.
I accept all kind of ideas you can think of, I’m very grateful if you know documentation that doesn’t deal with it with a translator.
Don’t train in multiple languages but use a translator
Train in multiple languages without the use of a translator
You seem to want to train the bot using one language only, but understand and speak multiple ones, correct me if I’m wrong.
Now, you said “having to train each language separately”. No need to do it separately if you chose to go with #2. You can train them all at once like in this example.
I understand that writing it in language only, means using something similar to google translator to adapt to the user’s language, correct me if I’m wrong? But that’s not what I’m looking for, I wouldn’t want the chatbot to work with translator.
I’m looking to do something more similar to the example you sent, but I see a problem, if there are few intents and entities nothing happens, but if there are many the code is multiplied/triplicated/ (depending on the number of languages). I understand that it is necessary to write these things but I don’t know if it is the only way and if it is the best way to do it this way.
Yup, every language you add will double the data again.
This is why approach #1 is a lot more scalable (just write the data in one language and you’re done), but approach #2 is a lot more controlled (no mistranslations).