app.py

import numpy as np
import gradio as gr


def textMode(text, count):
    return "In progress"


def imageMode(image, question):
    return "In progress"

def audioMode(audio):
    return "In progress"


interface_title = "TSAI-ERA-V1 - Capstone - Multimodal GPT Demo"
with gr.Blocks() as demo:
    with gr.Row():
        gr.Markdown(f"<h1>{interface_title}</h1>", element_id="title")
    gr.Markdown("Choose text mode/image mode/audio mode for generation")
    with gr.Tab("Text mode"):
        text_input = gr.Textbox(placeholder="Enter a prompt", label="Input")
        text_input_count = gr.Textbox(placeholder="Enter number of characters you want to generate", label="Count")
        text_button = gr.Button("Submit")
        text_output = gr.Textbox(label="Chat GPT like text")        
    with gr.Tab("Image mode"):
        with gr.Row():
            image_input = gr.Image()
            image_text_input = gr.Textbox(placeholder="Enter a question/prompt around the image", label="Question/Prompt")
        image_button = gr.Button("Submit")   
        image_text_output = gr.Textbox(label="Answer")
        
    with gr.Tab("Audio mode"):
        audio_input = gr.Audio()
        audio_button = gr.Button("Submit")
        audio_text_output = gr.Textbox(label="Chat GPT like text")
        

    text_button.click(textMode, inputs=[text_input, text_input_count], outputs=text_output)
    image_button.click(imageMode, inputs=[image_input,image_text_input], outputs=image_text_output)
    audio_button.click(audioMode, inputs=audio_input, outputs=audio_text_output)

demo.launch()