Update app.py

app.py

@@ -1,51 +1,74 @@
 import os
-os.environ["HF_HUB_DISABLE_TELEMETRY"] = "1"
-os.environ["HF_HUB_ENABLE_HF_TRANSFER"] = "1"
 import gradio as gr
 from multimodal_module import MultiModalChatModule
 import asyncio
 
-AI = MultiModalChatModule()
+# Initialize module
+mm = MultiModalChatModule()
 
-async def chat_async(text, user_id, lang):
-    return await AI.generate_response(text, int(user_id), lang)
+# Environment configuration
+os.environ["HF_HUB_DISABLE_TELEMETRY"] = "1"
+os.environ["TOKENIZERS_PARALLELISM"] = "false"
 
-def chat(text, user_id, lang="en"):
-    return asyncio.run(chat_async(text, user_id, lang))
+async def async_wrapper(fn, *args):
+    """Handle async calls from Gradio"""
+    return await fn(*args)
 
-async def voice_async(audio, user_id):
-    return await AI.process_voice_message(audio, int(user_id))
+def process_voice(audio, user_id):
+    return asyncio.run(async_wrapper(mm.process_voice_message, audio, int(user_id)))
 
-def voice(audio, user_id):
-    return asyncio.run(voice_async(audio, user_id))
+def process_image(image, user_id):
+    return asyncio.run(async_wrapper(mm.process_image_message, image, int(user_id)))
 
-async def image_async(image, user_id):
-    return await AI.process_image_message(image, int(user_id))
+def chat(text, user_id, lang):
+    return asyncio.run(async_wrapper(mm.generate_response, text, int(user_id), lang))
 
-def image(image, user_id):
-    return asyncio.run(image_async(image, user_id))
+def generate_image(prompt, user_id):
+    return asyncio.run(async_wrapper(
+        mm.generate_image_from_text, 
+        prompt, 
+        int(user_id)
+    ))
 
-with gr.Blocks() as demo:
-    gr.Markdown("# Multimodal Bot API")
+with gr.Blocks(title="Multimodal AI Assistant") as app:
+    gr.Markdown("## 🚀 Multimodal AI Assistant")
     
-    with gr.Tab("Text Chat"):
-        t_in = gr.Textbox(label="Message")
-        u_in = gr.Textbox(label="User ID", value="123")
-        l_in = gr.Textbox(label="Language", value="en")
-        t_out = gr.Textbox(label="AI Response")
-        gr.Button("Send").click(chat, inputs=[t_in, u_in, l_in], outputs=t_out)
+    with gr.Tab("💬 Text Chat"):
+        with gr.Row():
+            user_id = gr.Textbox(label="User ID", value="123")
+            lang = gr.Dropdown(["en", "es", "fr", "de"], label="Language", value="en")
+        chat_input = gr.Textbox(label="Your Message")
+        chat_output = gr.Textbox(label="AI Response", interactive=False)
+        chat_btn = gr.Button("Send")
     
-    with gr.Tab("Voice"):
-        v_in = gr.Audio(sources=["microphone", "upload"], type="filepath")
-        vu_in = gr.Textbox(label="User ID", value="123")
-        v_out = gr.JSON(label="Voice Analysis")
-        gr.Button("Process").click(voice, inputs=[v_in, vu_in], outputs=v_out)
+    with gr.Tab("🎙️ Voice"):
+        voice_input = gr.Audio(sources=["microphone", "upload"], type="filepath")
+        voice_user = gr.Textbox(label="User ID", value="123")
+        voice_output = gr.JSON(label="Analysis Results")
+        voice_btn = gr.Button("Process")
     
-    with gr.Tab("Image"):
-        i_in = gr.Image(type="filepath")
-        iu_in = gr.Textbox(label="User ID", value="123")
-        i_out = gr.Textbox(label="Image Description")
-        gr.Button("Describe").click(image, inputs=[i_in, iu_in], outputs=i_out)
-
-demo.queue()
-demo.launch()
+    with gr.Tab("🖼️ Images"):
+        with gr.Tab("Describe"):
+            img_input = gr.Image(type="filepath")
+            img_user = gr.Textbox(label="User ID", value="123")
+            img_output = gr.Textbox(label="Description")
+            img_btn = gr.Button("Describe")
+        
+        with gr.Tab("Generate"):
+            gen_prompt = gr.Textbox(label="Prompt")
+            gen_user = gr.Textbox(label="User ID", value="123")
+            gen_output = gr.Image(label="Generated Image")
+            gen_btn = gr.Button("Generate")
+    
+    # Event handlers
+    chat_btn.click(chat, [chat_input, user_id, lang], chat_output)
+    voice_btn.click(process_voice, [voice_input, voice_user], voice_output)
+    img_btn.click(process_image, [img_input, img_user], img_output)
+    gen_btn.click(generate_image, [gen_prompt, gen_user], gen_output)
+
+if __name__ == "__main__":
+    app.launch(
+        server_name="0.0.0.0",
+        server_port=7860,
+        share=False
+    )