Update app.py

app.py

@@ -3,6 +3,7 @@ from gtts import gTTS
 import time
 from fastapi import FastAPI, Request
 from fastapi.responses import HTMLResponse
+from fastapi.responses import FileResponse
 from fastapi.staticfiles import StaticFiles
 from llama_index.core import StorageContext, load_index_from_storage, VectorStoreIndex, SimpleDirectoryReader, ChatPromptTemplate, Settings
 from llama_index.llms.huggingface import HuggingFaceInferenceAPI
@@ -17,15 +18,18 @@ from huggingface_hub import InferenceClient
 import json
 import re
 from deep_translator import GoogleTranslator
+from dotenv import load_dotenv
+import random
+import string
 
 
-
+load_dotenv()
 # Define Pydantic model for incoming request body
 class MessageRequest(BaseModel):
     message: str
     language: str
 
-repo_id = "openai-community/gpt2"
+repo_id = "meta-llama/Meta-Llama-3-8B-Instruct"
 llm_client = InferenceClient(
     model=repo_id,
     token=os.getenv("HF_TOKEN"),
@@ -61,8 +65,8 @@ templates = Jinja2Templates(directory="static")
 
 # Configure Llama index settings
 Settings.llm = HuggingFaceInferenceAPI(
-    model_name="openai-community/gpt2",
-    tokenizer_name="openai-community/gpt2",
+    model_name="meta-llama/Meta-Llama-3-8B-Instruct",
+    tokenizer_name="meta-llama/Meta-Llama-3-8B-Instruct",
     context_window=3000,
     token=os.getenv("HF_TOKEN"),
     max_new_tokens=512,
@@ -86,7 +90,17 @@ def data_ingestion_from_directory():
     storage_context = StorageContext.from_defaults()
     index = VectorStoreIndex.from_documents(documents)
     index.storage_context.persist(persist_dir=PERSIST_DIR)
+def generate_unique_filename(extension="txt"):
+    # Current timestamp
 
+    timestamp = time.strftime("%Y%m%d%H%M%S")
+    
+    # Generate a random string of 6 characters
+    random_str = ''.join(random.choices(string.ascii_letters + string.digits, k=6))
+    
+    # Combine timestamp and random string
+    unique_filename = f"{timestamp}_{random_str}.{extension}"
+    return unique_filename
 def initialize():
     start_time = time.time()
     data_ingestion_from_directory()  # Process PDF ingestion at startup
@@ -110,13 +124,20 @@ def split_name(full_name):
     return first_name, last_name
 
 initialize()  # Run initialization tasks
-
+#You are the ITC GrandChola Hotel voice chatbot and your name is  hotel helper.
+#  Your goal is to provide accurate, professional, and helpful answers to user queries
+#  based on the Taj hotel data. Always ensure your responses are clear and concise. 
+# Give response within 10-15 words only. If you don't know the answer,
+#  you can say 'I don't know'. If you need more information, you can ask the user for clarification.
+#  You can also ask the user if they need help with anything else. Remember to be polite and professional at all times. If you are ready to start, you can say 'I am ready'. If you need to take a break, you can say 'I need a break'.
+#  If you need to end the conversation, you can say 'Goodbye'.
+            
 def handle_query(query):
     chat_text_qa_msgs = [
         (
             "user",
             """
-            You are the ITC GrandChola Hotel voice chatbot and your name is  hotel helper. Your goal is to provide accurate, professional, and helpful answers to user queries based on the Taj hotel data. Always ensure your responses are clear and concise. Give response within 10-15 words only. If you don't know the answer, you can say 'I don't know'. If you need more information, you can ask the user for clarification. You can also ask the user if they need help with anything else. Remember to be polite and professional at all times. If you are ready to start, you can say 'I am ready'. If you need to take a break, you can say 'I need a break'. If you need to end the conversation, you can say 'Goodbye'.
+            You are an assistant designed to help customers with inquiries about ITC Grand Chola Hotel. Your goal is to provide accurate, professional, and helpful answers based on available hotel information. If the user greets or doesn't ask a specific question, respond politely by encouraging them to ask about the hotel. Answer questions in a single sentence, within 5-10 words, while maintaining politeness and professionalism at all times.
             {context_str}
             Question:
             {query_str}
@@ -144,7 +165,8 @@ def handle_query(query):
         response = "Sorry, I couldn't find an answer."
     current_chat_history.append((query, response))
     return response
-
+def generate_unique_audio_filename():
+    return f"audio/response_{uuid.uuid4().hex}.mp3"
 @app.get("/ch/{id}", response_class=HTMLResponse)
 async def load_chat(request: Request, id: str):
     return templates.TemplateResponse("index.html", {"request": request, "user_id": id})
@@ -152,7 +174,13 @@ async def load_chat(request: Request, id: str):
 async def load_chat(request: Request, id: str):
     return templates.TemplateResponse("voice.html", {"request": request, "user_id": id})
 
-
+@app.get("/audio/{filename}")
+async def get_audio(filename: str):
+    audio_path = os.path.join(os.getcwd(), filename)  # Ensure correct file path
+    if os.path.exists(audio_path):
+        return FileResponse(audio_path)
+    else:
+        raise HTTPException(status_code=404, detail="Audio file not found.")
 
 @app.post("/chat/")
 async def chat(request: MessageRequest):
@@ -182,12 +210,14 @@ async def chat(request: MessageRequest):
     }
     chat_history.append(message_data)
     tts = gTTS(text=response1, lang=language_code)
-    audio_path = "response.mp3"
+    audio_path = generate_unique_filename("mp3")
     tts.save(audio_path)
     return {"response": response1,
-           "audioUrl": f"https://redfernstech-hotel-gtts.hf.space/audio/{audio_path}"}
+           "audioUrl": f"http://localhost:8000/audio/{audio_path}"}
+
 
 @app.get("/")
 def read_root(request: Request):
     return templates.TemplateResponse("home.html", {"request": request})
 
+