I have a use case I;e
I have 2 Bots, each Bot serves different functionalities and in different languages. So, I have created 3 RASA APPS and deployed them in a virtual machine and running in different ports.
HR BOT - It is just QnA, I have Created HR_RASA_APP Whenever a user asks a question, it returns the answer from RASA APP and that answer we are displaying in the Chatbot.
FINANCE BOT - It is more of a functional Bot, where we call our systems to get the data based on the INTENT and ENTITY identified. I have created FINANCE_RASA_APP In this APP, We just identify the INTENT and ENTITIES from the user message, Based on that we will do the remaining operations at the backend.
FINANCE BOT (GERMAN LANGUAGE) - It is the same functionality as the 2nd one. But we are supporting in GERMAN LANGUAGE.
Currently, what we did is, We have deployed 3 RASA APPs running on 3 different ports. But it’s not a good approach in the overall architecture point of view. So, we are trying to check different possibilities:
1. Can we run multiple RASA apps on a Single Port in the Same VM?
2. Can we use a Single Rasa app to support in different languages? Like when we Train the app, we will use one model for English and another Trained model for German. is that Possible?