Conversational Retrieval Chain #multi_prompts

[nedala10]

i am creating a chatbot by langchain so i am using a ConversationalRetrievalChain , so i want to determine some prompts to improve my output.

i want to give the bot name ,character and behave (system message prompt )

users use different languages how can i let the bot take user input then translate it to English then parse it with my data which is by English language, after getting the answer, the bot should pass the answer with same user language

[BioStarr]

from langchain_core.prompts.prompt import PromptTemplate


template = """You are chatbot your name is Tutakamon...etc
Question: {question}
 =========
 {context}
=========
"""

QA_PROMPT = PromptTemplate(template=template,input_variables=[
                           "question", "context","chat_history"])

qa = ConversationalRetrievalChain.from_llm(combine_docs_chain_kwargs={"prompt": QA_PROMPT})