app.py

import gradio as gr
from project_model import process_inputs


def handle_inputs(image, audio):
    if image is None or audio is None:
        return "Please upload both an image and an audio clip.", None, None
    message, answer_audio = process_inputs(image, audio)
    return message, image, answer_audio

with gr.Blocks() as demo:
    gr.Markdown("## 👁️🎙️ Multimodal Visual Q&A with Audio Output")

    with gr.Row():
        with gr.Column():
            image_input = gr.Image(label="Upload or Capture Image", sources=["upload", "webcam"], type="pil")
            audio_input = gr.Audio(label="Record Voice", sources=["microphone"], type="filepath")
            submit_btn = gr.Button("Submit")
        
        with gr.Column():
            status_output = gr.Textbox(label="Status", interactive=False)
            image_display = gr.Image(label="Processed Image")
            audio_output = gr.Audio(label="Answer Audio", interactive=False)

    submit_btn.click(fn=handle_inputs, inputs=[image_input, audio_input],
                     outputs=[status_output, image_display, audio_output])

if __name__ == "__main__":
    demo.launch(show_error=True, share=True)