app.py

import gradio as gr
import torch
from diffusers import AutoPipelineForText2Image
import base64
from io import BytesIO
from generate_propmts import generate_prompt
from concurrent.futures import ThreadPoolExecutor

# Load the model once outside of the function
model = AutoPipelineForText2Image.from_pretrained("stabilityai/sdxl-turbo")

def generate_image(text, sentence_mapping, character_dict, selected_style):
    try:
        prompt, _ = generate_prompt(text, sentence_mapping, character_dict, selected_style)
        image = model(prompt=prompt, num_inference_steps=1, guidance_scale=0.0).images[0]
        buffered = BytesIO()
        img_str = base64.b64encode(buffered.getvalue()).decode("utf-8")
        if isinstance(result, img_str):
            image_bytes = base64.b64decode(result)
        return image_bytes
    except Exception as e:
        print(f"Error generating image: {e}")
        return None

def inference(text, sentence_mapping, character_dict, selected_style):
    images = {}
    # Here we assume `sentence_mapping` is a dictionary where keys are paragraph numbers and values are lists of sentences
    grouped_sentences = sentence_mapping

    with ThreadPoolExecutor() as executor:
        futures = {}
        for paragraph_number, sentences in grouped_sentences.items():
            combined_sentence = " ".join(sentences)
            futures[paragraph_number] = executor.submit(generate_image, combined_sentence, sentence_mapping, character_dict, selected_style)

        for paragraph_number, future in futures.items():
            images[paragraph_number] = future.result()

    return images

gradio_interface = gr.Interface(
    fn=inference,
    inputs="text",
    outputs="text"  
)

if __name__ == "__main__":
    gradio_interface.launch()