Upload folder using huggingface_hub

.gradio/certificate.pem

@@ -0,0 +1,31 @@
+-----BEGIN CERTIFICATE-----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=
+-----END CERTIFICATE-----

README.md

@@ -1,12 +1,6 @@
 ---
 title: ConversAI
-emoji: ⚡
-colorFrom: gray
-colorTo: blue
+app_file: insight_genie_v021.py
 sdk: gradio
-sdk_version: 5.46.1
-app_file: app.py
-pinned: false
+sdk_version: 5.45.0
 ---
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

chat,py.py

@@ -0,0 +1,29 @@
+import gradio as gr
+import json
+import requests
+import time
+
+# Configuration for LM Studio API
+LM_STUDIO_API_URL = "http://localhost:1234/v1/chat/completions"
+
+# Advanced System Prompt for InsightGenie
+system_prompt_content = """
+You are InsightGenie, an AI-powered qualitative research assistant. Your purpose is to conduct a structured interview to deeply understand a user's experience with a specific topic.
+
+**Instructions:**
+1.  **Persona:** You are a professional, neutral, and empathetic research interviewer. Maintain a supportive and curious tone.
+2.  **Goal:** Your primary goal is to gather rich, detailed qualitative data. Ask open-ended questions that encourage detailed responses.
+3.  **Conversation Flow:**
+    - After each user response, analyze the sentiment and key topics.
+    - Based on your analysis, generate **one** follow-up question to probe deeper. Do not ask multiple questions.
+    - You must keep the conversation focused on the topic.
+4.  **Structured Output:** After each user turn, you must respond with a JSON object. The JSON should contain two fields:
+    - `next_question`: The text of your next question for the user.
+    - `summary`: A brief, neutral summary of the user's last response. This helps for later analysis.
+
+**Example JSON Response:**
+```json
+{
+  "next_question": "Can you tell me more about why that was your favorite part?",
+  "summary": "The user had a positive experience and liked the fast delivery."
+}

chat.py

@@ -0,0 +1,89 @@
+import gradio as gr
+import json
+import requests
+import time
+
+# Configuration for LM Studio API
+LM_STUDIO_API_URL = "http://192.168.1.245:1234/v1/chat/completions"
+# Make sure to replace this with your model name from LM Studio
+LM_MODEL_NAME = "google/gemma-3-27b"
+
+# Advanced System Prompt for InsightGenie
+system_prompt_content = """
+You are InsightGenie, an AI-powered qualitative research assistant. Your purpose is to conduct a structured interview to deeply understand a user's experience with a specific topic.
+
+**Instructions:**
+1.  **Persona:** You are a professional, neutral, and empathetic research interviewer. Maintain a supportive and curious tone.
+2.  **Goal:** Your primary goal is to gather rich, detailed qualitative data. Ask open-ended questions that encourage detailed responses.
+3.  **Conversation Flow:**
+    - After each user response, analyze the sentiment and key topics.
+    - Based on your analysis, generate **one** follow-up question to probe deeper. Do not ask multiple questions.
+    - You must keep the conversation focused on the topic.
+4.  **Structured Output:** After each user turn, you must respond with a JSON object. The JSON should contain two fields:
+    - `next_question`: The text of your next question for the user.
+    - `summary`: A brief, neutral summary of the user's last response. This helps for later analysis.
+
+**Example JSON Response:**
+```json
+{
+  "next_question": "Can you tell me more about why that was your favorite part?",
+  "summary": "The user had a positive experience and liked the fast delivery."
+}
+"""
+
+def chat_with_lm_studio(message, history):
+    messages = [{"role": "system", "content": system_prompt_content}]
+
+    for user_msg, assistant_msg in history:
+        messages.append({"role": "user", "content": user_msg})
+        messages.append({"role": "assistant", "content": assistant_msg})
+    
+    messages.append({"role": "user", "content": message})
+
+    try:
+        response = requests.post(
+            LM_STUDIO_API_URL,
+            json={
+                "model": LM_MODEL_NAME,
+                "messages": messages,
+                "max_tokens": 150,
+                "temperature": 0.7
+            }
+        )
+        response.raise_for_status()
+        
+        # Parse the JSON response
+        api_response_data = response.json()
+
+        # Check if 'choices' key exists in the response
+        if 'choices' in api_response_data and len(api_response_data['choices']) > 0:
+            ai_message_content = api_response_data['choices'][0]['message']['content']
+            
+            # Since we removed structured output, we just return the text
+            return ai_message_content
+        else:
+            # If 'choices' is missing, there's likely an error.
+            # Look for an 'error' key or other diagnostic info.
+            error_message = api_response_data.get('error', 'Unknown API error.')
+            print(f"API Error Response: {error_message}")
+            return f"An error occurred with the API: {error_message}"
+
+    except requests.exceptions.RequestException as e:
+        return f"An API error occurred: {e}. Please ensure LM Studio server is running."
+    except Exception as e:
+        return f"An unexpected error occurred: {e}"
+
+# Gradio interface to launch the demo
+demo = gr.ChatInterface(
+    fn=chat_with_lm_studio,
+    chatbot=gr.Chatbot(height=500),
+    theme=gr.themes.Soft(),
+    title="InsightGenie Live Demo",
+    description="Your AI-powered qualitative research assistant.",
+    examples=[["I had a great experience with a new online clothing store."], 
+              ["The delivery was slow, and the product was damaged."]],
+    multimodal=False
+)
+
+if __name__ == "__main__":
+    demo.launch(inbrowser=True)

conversation_log_1758405088.json

@@ -0,0 +1,14 @@
+[
+    {
+        "user_message": "Tell me about an experience with Metformin",
+        "ai_response": "Could you walk me through what your experience was like *immediately* after starting Metformin – within the first week or so?",
+        "ai_summary": "The user is being asked about the circumstances leading to their initiation of Metformin treatment.",
+        "timestamp": "2025-09-20 17:50:19"
+    },
+    {
+        "user_message": "It felt lightheaded",
+        "ai_response": "Can you describe what that feeling of lightheadedness was like in more detail? For example, did it come on suddenly, or was it a gradual sensation?",
+        "ai_summary": "The user reported experiencing lightheadedness upon starting Metformin.",
+        "timestamp": "2025-09-20 17:51:07"
+    }
+]

conversation_log_1758405091.json

@@ -0,0 +1,14 @@
+[
+    {
+        "user_message": "Tell me about an experience with Metformin",
+        "ai_response": "Could you walk me through what your experience was like *immediately* after starting Metformin – within the first week or so?",
+        "ai_summary": "The user is being asked about the circumstances leading to their initiation of Metformin treatment.",
+        "timestamp": "2025-09-20 17:50:19"
+    },
+    {
+        "user_message": "It felt lightheaded",
+        "ai_response": "Can you describe what that feeling of lightheadedness was like in more detail? For example, did it come on suddenly, or was it a gradual sensation?",
+        "ai_summary": "The user reported experiencing lightheadedness upon starting Metformin.",
+        "timestamp": "2025-09-20 17:51:07"
+    }
+]

insight_genie.py

@@ -0,0 +1,177 @@
+import gradio as gr
+import json
+import requests
+import time
+
+# --- Configuration for LM Studio API ---
+# Make sure your LM Studio server is running with the specified model
+LM_STUDIO_API_URL = "http://192.168.1.245:1234/v1/chat/completions"
+# Make sure to replace this with your model name from LM Studio exactly
+LM_MODEL_NAME = "google/gemma-3-27b"
+
+# --- Advanced System Prompt for Pharma Market Research ---
+system_prompt_content = """
+You are InsightGenie, an AI-powered qualitative research assistant specialized in pharmaceutical and healthcare market research. Your purpose is to conduct a structured interview to understand patient, caregiver, or healthcare professional (HCP) experiences with a specific health condition or treatment.
+
+**Instructions:**
+1. **Persona:** You are a professional, neutral, and empathetic research interviewer. Use clear, simple language when speaking with patients and caregivers, and appropriate medical terminology when speaking with HCPs. Maintain a supportive and curious tone.
+2. **Goal:** Your primary goal is to gather rich, detailed qualitative data. Ask open-ended questions that encourage detailed responses about personal experiences, emotional impact, and decision-making processes.
+3. **Compliance:** Avoid providing any medical advice, diagnoses, or treatment recommendations. State that you are a research tool and not a substitute for a healthcare professional.
+4. **Conversation Flow:**
+    - After each user response, analyze the sentiment and key themes.
+    - Based on your analysis, generate **one** follow-up question to probe deeper. Do not ask multiple questions.
+    - You must keep the conversation focused on the specified health topic.
+5. **Structured Output:** After each user turn, you must respond with a JSON object. The JSON should contain two fields:
+    - `next_question`: The text of your next question for the user.
+    - `summary`: A brief, neutral summary of the user's last response, including key terms or concepts.
+
+**Example JSON Response for a patient interview:**
+```json
+{
+  "next_question": "Can you describe the biggest challenges you faced when you were first diagnosed with this condition?",
+  "summary": "The patient shared their initial diagnosis experience, mentioning feelings of uncertainty."
+}
+"""
+
+# Global variable to store the conversation log for the current session
+conversation_log = []
+
+# --- Helper Functions ---
+
+def log_conversation_turn(user_message, ai_response, ai_summary):
+    """Appends a single turn to the in-memory conversation log."""
+    global conversation_log
+    conversation_log.append({
+        "user_message": user_message,
+        "ai_response": ai_response,
+        "ai_summary": ai_summary,
+        "timestamp": time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
+    })
+
+def save_conversation_log():
+    """Saves the entire conversation log to a JSON file."""
+    global conversation_log
+    if not conversation_log:
+        return "No conversation to save."
+    
+    file_name = f"conversation_log_{int(time.time())}.json"
+    try:
+        with open(file_name, 'w', encoding='utf-8') as f:
+            json.dump(conversation_log, f, indent=4, ensure_ascii=False)
+        return f"Conversation saved to {file_name}"
+    except Exception as e:
+        return f"Failed to save conversation: {e}"
+
+def start_new_session(chatbot_history):
+    """
+    Saves the current conversation and starts a new, empty session.
+    """
+    global conversation_log
+    
+    # Save the current conversation log
+    save_message = save_conversation_log()
+    
+    # Reset the in-memory log for the new session
+    conversation_log = []
+    
+    # Clear the Gradio chatbot history for a fresh start
+    return [], gr.Textbox(value=save_message, visible=True)
+
+# --- Core Chat Logic Function ---
+
+# --- Core Chat Logic Function ---
+def chat_with_lm_studio(message, history):
+    # This line ensures 'messages' is always defined at the start of the function.
+    messages = [{"role": "system", "content": system_prompt_content}]
+
+    for user_msg, assistant_msg in history:
+        messages.append({"role": "user", "content": user_msg})
+        messages.append({"role": "assistant", "content": assistant_msg})
+    
+    messages.append({"role": "user", "content": message})
+
+    # The rest of the function remains the same, using the now-defined 'messages' variable.
+    try:
+        # In your chat_with_lm_studio function
+# ...
+        response = requests.post(
+        LM_STUDIO_API_URL,
+        json={
+            "model": LM_MODEL_NAME,
+            "messages": messages,
+            "max_tokens": 150,
+            "temperature": 0.7,
+        # Remove or comment out this line:
+        # "response_format": {"type": "json_object"}
+    }
+)
+# ...
+        
+        response.raise_for_status()
+
+        api_response_data = response.json()
+        
+        if 'choices' in api_response_data and len(api_response_data['choices']) > 0:
+            raw_content = api_response_data['choices'][0]['message']['content']
+            
+            try:
+                parsed_response = json.loads(raw_content)
+                next_question = parsed_response.get("next_question", "Thank you for your response.")
+                summary = parsed_response.get("summary", "No summary provided.")
+                
+                log_conversation_turn(message, next_question, summary)
+                print(f"User: {message}\nAI Summary: {summary}\nAI Question: {next_question}\n---")
+
+                # The fix is here: Return both the user message and the AI response
+                history.append((message, next_question))
+                
+                # To clear the user input textbox, you need to return an empty string
+                return "", history
+
+            except json.JSONDecodeError:
+                print("LLM failed to produce valid JSON. Raw output:", raw_content)
+                history.append((message, "I'm sorry, I couldn't process that response. Can you please rephrase?"))
+                return "", history
+
+        else:
+            error_message = api_response_data.get('error', 'Unknown API error.')
+            print(f"API Error Response: {error_message}")
+            history.append((message, f"An error occurred with the API: {error_message}. Please check the console."))
+            return "", history
+
+    except requests.exceptions.RequestException as e:
+        history.append((message, f"An API error occurred: {e}. Please ensure LM Studio server is running and accessible."))
+        return "", history
+    except Exception as e:
+        history.append((message, f"An unexpected error occurred: {e}"))
+        return "", history
+# --- Gradio Interface Layout ---
+with gr.Blocks(theme=gr.themes.Soft(), title="InsightGenie Live Demo") as demo:
+    gr.Markdown("# InsightGenie: Your AI-powered Qualitative Assistant 🧠")
+    gr.Markdown(
+        "Start a conversation with our AI researcher. The conversation data is "
+        "automatically structured for analysis and can be saved to a file. "
+        "Try asking about a patient's journey or an HCP's experience with a treatment."
+    )
+    
+    # Textbox to display status messages (e.g., "Conversation saved!")
+    status_message = gr.Textbox(label="Status", interactive=False, visible=False)
+
+    chatbot = gr.Chatbot(height=500, placeholder="Type your first message to begin the interview...")
+    msg = gr.Textbox(label="Your message")
+
+    with gr.Row():
+        chat_submit_btn = gr.Button("Send")
+        chat_clear_btn = gr.Button("Clear Chat")
+        new_session_btn = gr.Button("Start New Session")
+
+    # Event handlers
+    msg.submit(chat_with_lm_studio, [msg, chatbot], [msg, chatbot], concurrency_limit=None)
+    chat_submit_btn.click(chat_with_lm_studio, [msg, chatbot], [msg, chatbot], concurrency_limit=None)
+    chat_clear_btn.click(lambda: [], None, [chatbot]) # Updated to correctly clear the chatbot history
+    
+    new_session_btn.click(start_new_session, [chatbot], [chatbot, status_message])
+
+# --- Launch the Demo ---
+if __name__ == "__main__":
+    demo.launch(inbrowser=True)

insight_genie_v02.py

@@ -0,0 +1,184 @@
+import gradio as gr
+import json
+import requests
+import time
+
+# --- Configuration for LM Studio API ---
+LM_STUDIO_API_URL = "http://192.168.1.245:1234/v1/chat/completions"
+LM_MODEL_NAME = "google/gemma-3-27b"
+
+# --- Dynamic System Prompt ---
+DEFAULT_PROMPT = """
+You are InsightGenie, an AI-powered qualitative research assistant. Your purpose is to conduct a structured interview to deeply understand a user's experience with a specific topic.
+
+**Instructions:**
+1. **Persona:** You are a professional, neutral, and empathetic research interviewer. Maintain a supportive and curious tone.
+2. **Goal:** Your primary goal is to gather rich, detailed qualitative data. Ask open-ended questions that encourage detailed responses.
+3. **Conversation Flow:**
+    - After each user response, analyze the sentiment and key topics.
+    - Based on your analysis, generate **one** follow-up question to probe deeper. Do not ask multiple questions.
+    - You must keep the conversation focused on the topic.
+4. **Structured Output:** After each user turn, you must respond with a JSON object. The JSON should contain two fields:
+    - `next_question`: The text of your next question for the user.
+    - `summary`: A brief, neutral summary of the user's last response.
+
+**Example JSON Response:**
+```json
+{
+  "next_question": "Can you tell me more about why that was your favorite part?",
+  "summary": "The user had a positive experience and liked the fast delivery."
+}
+"""
+
+# Global variable to store the conversation log for the current session
+conversation_log = []
+
+# --- Helper Functions ---
+def handle_save_and_display_status():
+    save_message = save_conversation_log()
+    # Returns a Gradio component update
+    return gr.Textbox(value=save_message, visible=True)
+
+def log_conversation_turn(user_message, ai_response, ai_summary):
+    """Appends a single turn to the in-memory conversation log."""
+    global conversation_log
+    conversation_log.append({
+        "user_message": user_message,
+        "ai_response": ai_response,
+        "ai_summary": ai_summary,
+        "timestamp": time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
+    })
+
+def save_conversation_log():
+    """Saves the entire conversation log to a JSON file."""
+    global conversation_log
+    if not conversation_log:
+        return "No conversation to save."
+    
+    file_name = f"conversation_log_{int(time.time())}.json"
+    try:
+        with open(file_name, 'w', encoding='utf-8') as f:
+            json.dump(conversation_log, f, indent=4, ensure_ascii=False)
+        return f"Conversation saved to {file_name}"
+    except Exception as e:
+        return f"Failed to save conversation: {e}"
+
+def chat_with_lm_studio(message, history, prompt_text):
+    messages = [{"role": "system", "content": prompt_text}]
+    for user_msg, assistant_msg in history:
+        messages.append({"role": "user", "content": user_msg})
+        messages.append({"role": "assistant", "content": assistant_msg})
+    messages.append({"role": "user", "content": message})
+
+    try:
+        response = requests.post(
+            LM_STUDIO_API_URL,
+            json={
+                "model": LM_MODEL_NAME,
+                "messages": messages,
+                "max_tokens": 250, # Increased max tokens to give the model more room
+                "temperature": 0.7
+            }
+        )
+        response.raise_for_status()
+
+        api_response_data = response.json()
+        
+        if 'choices' in api_response_data and len(api_response_data['choices']) > 0:
+            raw_content = api_response_data['choices'][0]['message']['content']
+            
+            # --- Robust JSON Extraction and Parsing Logic ---
+            try:
+                # Find the start and end of the JSON block
+                json_start = raw_content.find("```json")
+                if json_start != -1:
+                    json_end = raw_content.find("```", json_start + 1)
+                    if json_end != -1:
+                        # Extract the pure JSON string
+                        json_string = raw_content[json_start + 7:json_end].strip()
+                    else:
+                        # Fallback if the closing tag is missing
+                        json_string = raw_content[json_start + 7:].strip()
+                else:
+                    # Fallback to the entire response if no JSON block is found
+                    json_string = raw_content.strip()
+
+                parsed_response = json.loads(json_string)
+                next_question = parsed_response.get("next_question", "Thank you for your response.")
+                summary = parsed_response.get("summary", "No summary provided.")
+                
+                log_conversation_turn(message, next_question, summary)
+                print(f"User: {message}\nAI Summary: {summary}\nAI Question: {next_question}\n---")
+
+                history.append((message, next_question))
+                return "", history
+
+            except json.JSONDecodeError:
+                print("LLM failed to produce valid JSON. Raw output:", raw_content)
+                history.append((message, "I'm sorry, I couldn't process that. Can you please rephrase?"))
+                return "", history
+
+        else:
+            error_message = api_response_data.get('error', 'Unknown API error.')
+            print(f"API Error Response: {error_message}")
+            history.append((message, f"An API error occurred: {error_message}. Please check the console."))
+            return "", history
+
+    except requests.exceptions.RequestException as e:
+        history.append((message, f"An API error occurred: {e}. Please ensure LM Studio server is running."))
+        return "", history
+    except Exception as e:
+        history.append((message, f"An unexpected error occurred: {e}"))
+        return "", history
+
+# --- Gradio Interface Layout ---
+with gr.Blocks(theme=gr.themes.Soft(), title="InsightGenie Live Demo") as demo:
+    gr.Markdown("# InsightGenie: Your AI-powered Qualitative Assistant 🧠")
+    
+    with gr.Tabs():
+        with gr.Tab("Live Demo"):
+            gr.Markdown(
+                "Start a conversation with the AI researcher. "
+                "The conversation data is structured for analysis and can be saved."
+            )
+            chatbot = gr.Chatbot(height=500, placeholder="Type your first message to begin the interview...")
+            
+            with gr.Row():
+                msg = gr.Textbox(label="Your message", scale=4)
+                chat_submit_btn = gr.Button("Send", scale=1)
+
+            gr.Examples(
+                examples=[
+                    ["I had a great experience with a new online clothing store."],
+                    ["The delivery was slow, and the product was damaged."]
+                ],
+                inputs=msg
+            )
+            
+            with gr.Row():
+                clear_btn = gr.Button("Clear Chat")
+                save_btn = gr.Button("Save Conversation")
+            
+            save_status = gr.Textbox(label="Save Status", interactive=False, visible=False)
+
+        with gr.Tab("Prompt Settings"):
+            gr.Markdown(
+                "Customize the AI's persona and instructions. "
+                "Changing this prompt will affect the next conversation turn."
+            )
+            prompt_input = gr.Textbox(
+                label="System Prompt",
+                value=DEFAULT_PROMPT,
+                lines=20,
+                interactive=True
+            )
+
+    # Event Handlers
+    msg.submit(chat_with_lm_studio, [msg, chatbot, prompt_input], [msg, chatbot], concurrency_limit=None)
+    chat_submit_btn.click(chat_with_lm_studio, [msg, chatbot, prompt_input], [msg, chatbot], concurrency_limit=None)
+    clear_btn.click(lambda: [], None, [chatbot])
+    save_btn.click(handle_save_and_display_status, None, save_status)
+
+# --- Launch the Demo ---
+if __name__ == "__main__":
+    demo.launch(share=True)

insight_genie_v021.py

@@ -0,0 +1,194 @@
+import gradio as gr
+import json
+import requests
+import time
+
+# --- Configuration for LM Studio API ---
+LM_STUDIO_API_URL = "http://192.168.1.245:1234/v1/chat/completions"
+LM_MODEL_NAME = "google/gemma-3-27b"
+
+# --- Dynamic System Prompt ---
+DEFAULT_PROMPT = """
+You are InsightGenie, an AI-powered qualitative research assistant. Your purpose is to conduct a structured interview to deeply understand a user's experience with a specific topic.
+
+**Instructions:**
+1. **Persona:** You are a professional, neutral, and empathetic research interviewer. Maintain a supportive and curious tone.
+2. **Goal:** Your primary goal is to gather rich, detailed qualitative data. Ask open-ended questions that encourage detailed responses.
+3. **Conversation Flow:**
+    - After each user response, analyze the sentiment and key topics.
+    - Based on your analysis, generate **one** follow-up question to probe deeper. Do not ask multiple questions.
+    - You must keep the conversation focused on the topic.
+4. **Structured Output:** After each user turn, you must respond with a JSON object. The JSON should contain two fields:
+    - `next_question`: The text of your next question for the user.
+    - `summary`: A brief, neutral summary of the user's last response.
+
+**Example JSON Response:**
+```json
+{
+  "next_question": "Can you tell me more about why that was your favorite part?",
+  "summary": "The user had a positive experience and liked the fast delivery."
+}
+"""
+
+# Global variable to store the conversation log for the current session
+conversation_log = []
+
+# --- Helper Functions ---
+def handle_save_and_display_status():
+    save_message = save_conversation_log()
+    return gr.Textbox(value=save_message, visible=True)
+
+def log_conversation_turn(user_message, ai_response, ai_summary):
+    """Appends a single turn to the in-memory conversation log."""
+    global conversation_log
+    conversation_log.append({
+        "user_message": user_message,
+        "ai_response": ai_response,
+        "ai_summary": ai_summary,
+        "timestamp": time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
+    })
+
+def save_conversation_log():
+    """Saves the entire conversation log to a JSON file."""
+    global conversation_log
+    if not conversation_log:
+        return "No conversation to save."
+    
+    file_name = f"conversation_log_{int(time.time())}.json"
+    try:
+        with open(file_name, 'w', encoding='utf-8') as f:
+            json.dump(conversation_log, f, indent=4, ensure_ascii=False)
+        return f"Conversation saved to {file_name}"
+    except Exception as e:
+        return f"Failed to save conversation: {e}"
+
+def chat_with_lm_studio(message, history, prompt_text):
+    messages = [{"role": "system", "content": prompt_text}]
+    for user_msg, assistant_msg in history:
+        messages.append({"role": "user", "content": user_msg})
+        messages.append({"role": "assistant", "content": assistant_msg})
+    messages.append({"role": "user", "content": message})
+
+    try:
+        response = requests.post(
+            LM_STUDIO_API_URL,
+            json={
+                "model": LM_MODEL_NAME,
+                "messages": messages,
+                "max_tokens": 250,
+                "temperature": 0.7
+            }
+        )
+        response.raise_for_status()
+
+        api_response_data = response.json()
+        
+        if 'choices' in api_response_data and len(api_response_data['choices']) > 0:
+            raw_content = api_response_data['choices'][0]['message']['content']
+            
+            try:
+                json_start = raw_content.find("```json")
+                if json_start != -1:
+                    json_end = raw_content.find("```", json_start + 1)
+                    if json_end != -1:
+                        json_string = raw_content[json_start + 7:json_end].strip()
+                    else:
+                        json_string = raw_content[json_start + 7:].strip()
+                else:
+                    json_string = raw_content.strip()
+
+                parsed_response = json.loads(json_string)
+                next_question = parsed_response.get("next_question", "Thank you for your response.")
+                summary = parsed_response.get("summary", "No summary provided.")
+                
+                log_conversation_turn(message, next_question, summary)
+                
+                transcript_message = (
+                    f"--- New Turn ---\n"
+                    f"User Input: {message}\n"
+                    f"AI Summary: {summary}\n"
+                    f"AI Question: {next_question}\n"
+                )
+                print(transcript_message)
+
+                history.append((message, next_question))
+                return "", history, transcript_message
+
+            except json.JSONDecodeError:
+                error_message = f"LLM failed to produce valid JSON. Raw output:\n{raw_content}"
+                print(error_message)
+                history.append((message, "I'm sorry, I couldn't process that. Can you please rephrase?"))
+                return "", history, error_message
+        
+        else:
+            error_message = api_response_data.get('error', 'Unknown API error.')
+            print(f"API Error Response: {error_message}")
+            history.append((message, f"An API error occurred: {error_message}. Please check the console."))
+            return "", history, f"API Error: {error_message}"
+
+    except requests.exceptions.RequestException as e:
+        history.append((message, f"An API error occurred: {e}. Please ensure LM Studio server is running."))
+        return "", history, f"API Error: {e}"
+    except Exception as e:
+        history.append((message, f"An unexpected error occurred: {e}"))
+        return "", history, f"Unexpected Error: {e}"
+
+# --- Gradio Interface Layout ---
+with gr.Blocks(theme=gr.themes.Soft(), title="Project Echo Live Demo") as demo:
+    gr.Markdown("# Project Echo: Your AI-powered Qualitative Assistant")
+    
+    with gr.Tabs():
+        with gr.Tab("Live Demo"):
+            gr.Markdown(
+                "Start a conversation with the AI researcher. "
+                "The conversation data is structured for analysis and can be saved."
+            )
+            chatbot = gr.Chatbot(height=500, placeholder="Start by telling me about your experience with a specific medication.")
+            
+            with gr.Row():
+                msg = gr.Textbox(label="Your message", scale=4)
+                chat_submit_btn = gr.Button("Send", scale=1)
+
+            gr.Examples(
+                examples=[
+                    ["What is a typical day like?"],
+                    ["What was your biggest challenge when first taking drug X?"]
+                ],
+                inputs=msg
+            )
+            
+            with gr.Row():
+                clear_btn = gr.Button("Clear Chat")
+                save_btn = gr.Button("Save Conversation")
+            
+            save_status = gr.Textbox(label="Save Status", interactive=False, visible=False)
+
+        with gr.Tab("Prompt Settings"):
+            gr.Markdown(
+                "Customize the AI's persona and instructions. "
+                "Changing this prompt will affect the next conversation turn."
+            )
+            prompt_input = gr.Textbox(
+                label="System Prompt",
+                value=DEFAULT_PROMPT,
+                lines=20,
+                interactive=True
+            )
+        
+        with gr.Tab("Unedited Interaction"):
+            log_output = gr.Textbox(
+                label="Backend Transcript",
+                interactive=False,
+                lines=15,
+                placeholder="Backend logs will appear here after each message is sent."
+            )
+
+    # Event Handlers
+    msg.submit(chat_with_lm_studio, [msg, chatbot, prompt_input], [msg, chatbot, log_output], concurrency_limit=None)
+    chat_submit_btn.click(chat_with_lm_studio, [msg, chatbot, prompt_input], [msg, chatbot, log_output], concurrency_limit=None)
+    clear_btn.click(lambda: [], None, [chatbot])
+    save_btn.click(handle_save_and_display_status, None, save_status)
+
+# --- Launch the Demo ---
+if __name__ == "__main__":
+    demo.launch(share=True)