Update app.py

app.py

@@ -38,17 +38,92 @@ transcriber = pipeline("automatic-speech-recognition", model="openai/whisper-bas
 # Set API Key (Ensure it's stored securely in an environment variable)
 groq.api_key = os.getenv("GROQ_API_KEY")  # Replace with a valid API key
 
+#___________________________________
+
+# Authenticate with Hugging Face API using the token
+#hf_token = os.getenv("HF_TOKEN")  # Replace with the environment variable containing your Hugging Face token
+#login(token=hf_token)
+
+# Load the LLaVA model
+#model_id = "liuhaotian/LLaVA-7B"  # You can change the model ID based on what is available
+#processor = AutoProcessor.from_pretrained(model_id)
+#model = AutoModelForVision2Seq.from_pretrained(model_id, torch_dtype=torch.float16).cuda()
+
+# Load and preprocess an image
+#image = Image.open("your_image.jpg")  # Replace with the path to your image
+#inputs = processor(text="Describe this image", images=image, return_tensors="pt").to("cuda")
+
+# Generate output from LLaVA model
+#output = model.generate(**inputs)
+
+# Decode and print the output
+#print(processor.decode(output[0]))
+
+#___________________________________
+
 # Initialize Chat Model
 chat_model = ChatGroq(model_name="llama-3.3-70b-versatile", api_key=groq.api_key) #DeepSeek-R1-Distill-Llama-70b , llama-3.3-70b-versatile , deepseek-r1-distill-qwen-32b
 
+# Initialize Embeddings and chromaDB
+
+embedding_model = HuggingFaceEmbeddings()
+vectorstore = Chroma(embedding_function=embedding_model)
+
+
+# -*- coding: utf-8 -*-
+"""app
+
+Automatically generated by Colab.
+
+Original file is located at
+    https://colab.research.google.com/drive/1jdKA4WQJcLb0_aQ3vtGVM46B1wriSsDv
+"""
+
+import gradio as gr
+import numpy as np
+from transformers import pipeline
+import os
+import time
+import groq
+import uuid  # For generating unique filenames
+
+# Updated imports to address LangChain deprecation warnings:
+from langchain_groq import ChatGroq
+from langchain.schema import HumanMessage
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_community.vectorstores import Chroma
+from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain.docstore.document import Document
+
+# Importing chardet (make sure to add chardet to your requirements.txt)
+import chardet
+
+import fitz  # PyMuPDF for PDFs
+import docx  # python-docx for Word files
+import gtts  # Google Text-to-Speech library
+from pptx import Presentation  # python-pptx for PowerPoint files
+import re
+
+
+# Initialize Whisper model for speech-to-text
+transcriber = pipeline("automatic-speech-recognition", model="openai/whisper-base.en")
+
+# Set API Key (Ensure it's stored securely in an environment variable)
+groq.api_key = os.getenv("GROQ_API_KEY")
+
+# Initialize Chat Model
+chat_model = ChatGroq(model_name="llama-3.3-70b-versatile", api_key=groq.api_key) #DeepSeek-R1-Distill-Llama-70b    |   deepseek-r1-distill-qwen-32b
+
 # Initialize Embeddings and chromaDB
 os.makedirs("chroma_db", exist_ok=True)
 embedding_model = HuggingFaceEmbeddings()
+#new
 vectorstore = Chroma(
     embedding_function=embedding_model,
     persist_directory="chroma_db"  # Set a valid folder name or path
 )
 vectorstore.persist()
+#end New
 
 # Short-term memory for the LLM
 chat_memory = []
@@ -96,7 +171,7 @@ d) 0.4
 
 Answer: d) 0.4
 
-Feedback: This question tests understanding of Bayes' Theorem by requiring the calculation of conditional probability using the given values.
+Feedback: This question tests understanding of Bayes’ Theorem by requiring the calculation of conditional probability using the given values.
 """
 
 # Function to clean AI response by removing unwanted formatting
@@ -122,7 +197,7 @@ def retrieve_documents(query):
     return [doc.page_content for doc in results]
 
 # Function to handle chatbot interactions with short-term memory
-def chat_with_groq(user_input, chat_history):
+def chat_with_groq(user_input):
     try:
         # Retrieve relevant documents for additional context
         relevant_docs = retrieve_documents(user_input)
@@ -143,16 +218,14 @@ def chat_with_groq(user_input, chat_history):
         chat_memory.append(f"User: {user_input}")
         chat_memory.append(f"AI: {cleaned_response_text}")
 
-        # Update chat history for Gradio
-        chat_history.append((user_input, cleaned_response_text))
-
         # Convert response to speech
         audio_file = speech_playback(cleaned_response_text)
 
-        return chat_history, "", audio_file
+        # Ensure the return format is a list of tuples
+        return [(user_input, cleaned_response_text)], audio_file
     except Exception as e:
-        chat_history.append((user_input, f"Error: {str(e)}"))
-        return chat_history, "", None
+        return [("Error", str(e))], None
+
 
 # Function to play response as speech using gTTS
 def speech_playback(text):
@@ -248,10 +321,37 @@ def transcribe_audio(audio):
     y /= np.max(np.abs(y))
     return transcriber({"sampling_rate": sr, "raw": y})["text"]
 
-# Clear chat history function
-def clear_chat_history():
-    chat_memory.clear()
-    return [], None
+# Modify chat_with_groq function to return audio file for playback
+def chat_with_groq(user_input):
+    try:
+        # Retrieve relevant documents for additional context
+        relevant_docs = retrieve_documents(user_input)
+        context = "\n".join(relevant_docs) if relevant_docs else "No relevant documents found."
+
+        # Construct proper prompting with conversation history
+        system_prompt = "You are a helpful AI assistant. Answer questions accurately and concisely."
+        conversation_history = "\n".join(chat_memory[-10:])  # Keep the last 10 exchanges
+        prompt = f"{system_prompt}\n\nConversation History:\n{conversation_history}\n\nUser Input: {user_input}\n\nContext:\n{context}"
+
+        # Call the chat model
+        response = chat_model([HumanMessage(content=prompt)])
+
+        # Clean response to remove any unwanted formatting
+        cleaned_response_text = clean_response(response.content)
+
+        # Append conversation history
+        chat_memory.append(f"User: {user_input}")
+        chat_memory.append(f"AI: {cleaned_response_text}")
+
+        # Convert response to speech
+        audio_file = speech_playback(cleaned_response_text)
+
+        # Return both chat response and audio file path
+        return [(user_input, cleaned_response_text)], audio_file  # Return as a tuple
+    except Exception as e:
+        return [("Error", str(e))], None
+
+#__________________________________________________________________________________________________________________________
 
 def tutor_ai_chatbot():
     """Main Gradio interface for the Tutor AI Chatbot."""
@@ -263,14 +363,17 @@ def tutor_ai_chatbot():
         with gr.Tab("AI Chatbot"):
             with gr.Row():
                 with gr.Column(scale=3):
-                    chatbot = gr.Chatbot(height=500, type="messages")
+                    chatbot = gr.Chatbot(height=500)  # Chatbot display area
+                    #chatbot = gr.Chatbot(height=500, type="messages")
                     with gr.Row():
                         msg = gr.Textbox(label="Ask a question", placeholder="Type your question here...")
                         submit = gr.Button("Send")
 
+                    #with gr.Row():
                     with gr.Column(scale=1):
                         audio_input = gr.Audio(type="numpy", label="Record or Upload Audio")   # Audio input for speech-to-text
 
+
                 with gr.Column(scale=1):
                    audio_playback = gr.Audio(label="Audio Response", type="filepath")
 
@@ -280,18 +383,21 @@ def tutor_ai_chatbot():
             # Handle chat interaction
             submit.click(
                 chat_with_groq,
-                inputs=[msg, chatbot],
-                outputs=[chatbot, msg, audio_playback]
+                inputs=[msg],
+                outputs=[chatbot, audio_playback]
             )
 
             # Clear chat history function
-            clear_btn.click(clear_chat_history, inputs=None, outputs=[chatbot, audio_playback])
+            def clear_chat_history():
+                return None, None
+
+            clear_btn.click(clear_chat_history, inputs=None, outputs=[chatbot, audio_playback])    #,audio_input
 
             # Also allow Enter key to submit
             msg.submit(
                 chat_with_groq,
-                inputs=[msg, chatbot],
-                outputs=[chatbot, msg, audio_playback]
+                inputs=[msg],
+                outputs=[chatbot, audio_playback]
             )
 
             # Add some examples of questions students might ask
@@ -307,30 +413,35 @@ def tutor_ai_chatbot():
                     inputs=msg
                 )
 
-            # Connect audio input to transcription
-            audio_input.change(fn=transcribe_audio, inputs=audio_input, outputs=msg)
-
         # Upload Notes & Generate Quiz Tab
         with gr.Tab("Upload Notes & Generate Quiz"):
             with gr.Row():
                 with gr.Column(scale=2):
                     file_input = gr.File(label="Upload Lecture Notes (PDF, DOCX, PPTX)")
+                    #generate_btn = gr.Button("Generate Quiz")
                 with gr.Column(scale=3):
                     quiz_output = gr.Textbox(label="Generated Quiz", lines=10)
 
-            # Connect file input to document processing
-            file_input.change(process_document, inputs=file_input, outputs=quiz_output)
 
          # Introduction Video
         with gr.Tab("Introduction Video"):
             with gr.Row():
                 with gr.Column(scale=1):
-                    gr.Markdown("### Welcome to the Introduction Video")
-                    gr.Markdown("Music from Xu Mengyuan - China-O, musician Xu Mengyuan YUAN! | 徐梦圆 - China-O 音乐人徐梦圆YUAN! ")
-                    gr.Video("https://huggingface.co/spaces/Lesterchia174/FPOC2_AI-Tutor_Chatbot/raw/main/We%20not%20me%20video.mp4", label="Introduction Video")
+                    #with gr.Column(scale=1):  # Adjust scale for equal width
+                    gr.Markdown("### Welcome to the Introduction Video")  # Adding a heading
+                    gr.Markdown("Music from Xu Mengyuan - China-O, musician Xu Mengyuan YUAN! | 徐梦圆 - China-O 音乐人徐梦圆YUAN! ")  # Adding descriptive text
+                    gr.Video("https://github.com/lesterchia1/AI_tutor/raw/main/We%20not%20me%20video.mp4", label="Introduction Video")
+                    #gr.Video("https://huggingface.co/spaces/Lesterchia174/FPOC2_AI-Tutor_Chatbot/raw/main/We%20not%20me%20video.mp4", label="Introduction Video")
+
+
+            # Connect the button to the document processing function
+            audio_input.change(fn=transcribe_audio, inputs=audio_input, outputs=msg)  # transcribe and fill the msg textbox
+            file_input.change(process_document, inputs=file_input, outputs=quiz_output)
+
 
     # Launch the application
-    app.launch(share=True)
+    app.launch(share=True)  # Set share=True to create a public link
+
 
 # Launch the AI chatbot
 if __name__ == "__main__":