import gradio as gr
from transformers import (
    AutoProcessor,
    VisionEncoderDecoderModel,
    ViTImageProcessor,
    AutoTokenizer,
    MusicgenForConditionalGeneration,
)
import torch
from PIL import Image
from openai import OpenAI
from scipy.io import wavfile
import numpy as np
import pic2song
import uuid


def greet(image, image_ins):
    file = pic2song.pic2song(image, image_ins)
    return file


with gr.Blocks() as demo:
    with gr.Column():
        with gr.Row():
            inp = gr.Image(type="pil", image_mode="RGB", height="500px")
            with gr.Column():
                image_ins = gr.Slider(
                    minimum=1,
                    maximum=60,
                    step=1,
                    value=5,
                    label="Generating length (seconds) 生成长度(秒)",
                )
                with gr.Row():
                    out1 = gr.Audio()
                    out2 = gr.Audio()
                with gr.Row():
                    out3 = gr.Audio()
                    out4 = gr.Audio()
    btn = gr.Button("Run")
    with gr.Column():
        with gr.Row():
            video_out1 = gr.Video(interactive=False,height="300px",show_download_button=True)
   
            video_out2 = gr.Video(interactive=False,height="300px",show_download_button=True)

            video_out3 = gr.Video(interactive=False,height="300px",show_download_button=True)

            video_out4 = gr.Video(interactive=False,height="300px",show_download_button=True)
    
    btn.click(
        fn=greet,
        inputs=[inp, image_ins],
        outputs=[
            out1,
            video_out1,
            out2,
            video_out2,
            out3,
            video_out3,
            out4,
            video_out4,
        ],
    )

demo.launch()