# import numpy as np
# from sklearn.metrics.pairwise import cosine_similarity
# from utils.convert_embedding import GetEmbedding
# import random
# import pickle
# import os 
# from utils.rag import RAG
# from faster_whisper import WhisperModel


# def process(user_query:str):
#     # dump_user_question(user_query)
#     user_embedding = GetEmbedding([user_query]).user_query_emb()
#     with open(r"all_mix_embedding.pkl","rb") as f:
#         load_embedding = pickle.load(f)

#     with open(r"all_answers.pkl","rb") as f:
#         ans = pickle.load(f)
#     similarity_scores = cosine_similarity(user_embedding, load_embedding)
#     index = np.argmax(similarity_scores)
#     answer = ans[index]
#     score = similarity_scores[0,index]
#     print(f"Index : {index}:\tscore:{score} \tquery: {user_query}")

#     if float(score) > 0.60 :
#         final_output = random.choice(answer)
#     else:
#         final_output = RAG().pipeline(query=user_query)

#     return final_output


# def audio_process(audio):
#     try:
#         model = WhisperModel(model_size_or_path="medium.en")
#         segments, info = model.transcribe(audio)
#         transcription = " ".join([seg.text for seg in segments])
#         result = process(user_query=transcription)
#         return result
#     except Exception as e:
#         print("Error:", e)
#         return str(e)




# if __name__ == "__main__":
#     res = audio_process(r"C:\Users\lalit\Documents\Sound recordings\who_is_lalit.m4a")
#     print(res)
#     # for _ in range(3):
    
#     #     user = input("How can i help you :? \n")
#     #     result = process(user)
#     #     print(result)

#     # with open(r"data\question_data.pkl","rb") as f:
#     #     que = pickle.load(f)
#     # print(que)