Update app.py

app.py

@@ -1,214 +1,32 @@
 import gradio as gr
 from transformers import pipeline
-import whisper
 
-# Load Qwen model
-chatbot = pipeline(
-    "text-generation",
-    model="Qwen/Qwen2.5-0.5B-Instruct",
-    trust_remote_code=True
-)
+chatbot = pipeline("text-generation", model="Qwen/Qwen2.5-0.5B-Instruct", trust_remote_code=True)
 
-# Load Whisper model
-whisper_model = whisper.load_model("base")
-
-def transcribe_audio(audio_file):
-    """Convert speech to text using Whisper"""
-    if audio_file is None:
-        return ""
-    
-    try:
-        result = whisper_model.transcribe(audio_file)
-        return result["text"].strip()
-    except Exception as e:
-        print(f"Transcription error: {e}")
-        return ""
-
-def get_ai_response(message):
-    """Get response from Qwen model"""
-    try:
-        prompt = f"Human: {message}\nAssistant:"
-        
-        response = chatbot(
-            prompt,
-            max_new_tokens=150,
-            temperature=0.7,
-            do_sample=True
-        )
-        
-        full_text = response[0]['generated_text']
-        assistant_response = full_text.split("Assistant:")[-1].strip()
-        
-        return assistant_response if assistant_response else "I'm thinking..."
-        
-    except Exception as e:
-        return f"Error: {str(e)[:100]}"
-
-def process_voice_input(audio, chat_history):
-    """Process voice input - FIXED message format"""
-    if audio is None:
-        return chat_history, "", None
-    
-    # Transcribe audio to text
-    text = transcribe_audio(audio)
-    
-    if not text:
-        # CORRECT FORMAT: list of dicts
-        chat_history.append({"role": "user", "content": "Voice input"})
-        chat_history.append({"role": "assistant", "content": "Sorry, I couldn't understand the audio."})
-        return chat_history, "", None
-    
-    # Get AI response
-    response = get_ai_response(text)
+def respond(message, history):
+    """Simplified response function"""
+    prompt = ""
+    for msg in history:
+        prompt += f"{msg['role'].capitalize()}: {msg['content']}\n"
     
-    # CORRECT FORMAT: list of dicts
-    chat_history.append({"role": "user", "content": f"🎤 {text}"})
-    chat_history.append({"role": "assistant", "content": response})
+    prompt += f"Human: {message}\nAssistant:"
     
-    return chat_history, "", None
+    response = chatbot(prompt, max_new_tokens=100)[0]['generated_text']
+    return response.split("Assistant:")[-1].strip()
 
-def process_text_input(text, chat_history):
-    """Process text input - FIXED message format"""
-    if not text.strip():
-        return chat_history, ""
+# Simple interface with voice
+with gr.Blocks() as demo:
+    gr.Markdown("# 🎤 Voice AI Assistant")
     
-    # Get AI response
-    response = get_ai_response(text)
-    
-    # CORRECT FORMAT: list of dicts
-    chat_history.append({"role": "user", "content": text})
-    chat_history.append({"role": "assistant", "content": response})
-    
-    return chat_history, ""
-
-def clear_chat():
-    """Clear chat history"""
-    return [], []
-
-def set_voice_mode():
-    """Set voice mode status"""
-    return "**Status:** Voice mode active - Click microphone to speak"
-
-def set_text_mode():
-    """Set text mode status"""
-    return "**Status:** Text mode active - Type your message"
-
-# Create the interface
-with gr.Blocks(fill_height=True) as demo:
-    
-    # Simple background animation
-    gr.HTML("""
-    <div style="position:fixed; top:0; left:0; width:100vw; height:100vh; 
-                background: linear-gradient(45deg, #0a0a0f, #1a1a2e); 
-                z-index:-1;">
-    </div>
-    """)
-    
-    # Main content
-    with gr.Column(elem_id="main-content"):
-        gr.Markdown("# 🎤 Voice-Enabled AI Assistant")
-        gr.Markdown("Talk to the AI using your voice or type your message")
-        
-        # Chat display
-        chatbot_display = gr.Chatbot(
-            label="Conversation",
-            height=400
-        )
-        
-        # Voice input section
-        with gr.Row():
-            with gr.Column(scale=1):
-                audio_input = gr.Audio(
-                    sources=["microphone"],
-                    type="filepath",
-                    label="🎤 Speak your message",
-                    interactive=True
-                )
-            
-            with gr.Column(scale=1):
-                gr.Markdown("### Or type your message:")
-                text_input = gr.Textbox(
-                    label="Type here",
-                    placeholder="Type your message...",
-                    lines=2
-                )
-                text_submit = gr.Button("Send Text", variant="primary")
-        
-        # Control buttons
-        with gr.Row():
-            clear_btn = gr.Button("Clear Chat", variant="secondary")
-            voice_mode_btn = gr.Button("🎤 Voice Mode", variant="primary")
-            text_mode_btn = gr.Button("📝 Text Mode", variant="secondary")
-        
-        # Status indicator
-        status = gr.Markdown("**Status:** Ready - Click microphone or type to chat")
-        
-        # State - initialize with welcome message
-        chat_history = gr.State([
-            {"role": "assistant", "content": "Hello! I'm your AI assistant. Speak into the microphone or type your message."}
-        ])
-    
-    # Event handlers
-    # Voice input
-    audio_input.stop_recording(
-        fn=process_voice_input,
-        inputs=[audio_input, chat_history],
-        outputs=[chatbot_display, text_input, audio_input]
-    )
-    
-    # Text input via button
-    text_submit.click(
-        fn=process_text_input,
-        inputs=[text_input, chat_history],
-        outputs=[chatbot_display, text_input]
-    )
-    
-    # Text input via Enter key
-    text_input.submit(
-        fn=process_text_input,
-        inputs=[text_input, chat_history],
-        outputs=[chatbot_display, text_input]
-    )
-    
-    # Clear chat
-    clear_btn.click(
-        fn=clear_chat,
-        outputs=[chatbot_display, chat_history]
+    # Chat interface
+    chat_interface = gr.ChatInterface(
+        fn=respond,
+        title="AI Chat",
+        description="Type or use voice input"
     )
     
-    # Mode buttons
-    voice_mode_btn.click(
-        fn=set_voice_mode,
-        outputs=[status]
-    )
-    
-    text_mode_btn.click(
-        fn=set_text_mode,
-        outputs=[status]
-    )
+    # Add voice input
+    with gr.Row():
+        audio = gr.Audio(sources=["microphone"], type="filepath", label="Voice Input")
 
-# Launch the app
-if __name__ == "__main__":
-    demo.launch(
-        server_name="0.0.0.0",
-        server_port=7860,
-        share=False,
-        theme=gr.themes.Soft(),
-        css="""
-        #main-content {
-            background: rgba(15, 15, 25, 0.85);
-            backdrop-filter: blur(10px);
-            padding: 30px;
-            border-radius: 20px;
-            border: 1px solid rgba(255, 255, 255, 0.1);
-            max-width: 800px;
-            margin: 50px auto;
-            box-shadow: 0 20px 40px rgba(0, 0, 0, 0.3);
-        }
-        
-        .gradio-container {
-            background: transparent !important;
-            min-height: 100vh !important;
-        }
-        """
-    )
+demo.launch()