app.py

import spaces
import gradio as gr
from wtpsplit import SaT
import json
import zipfile
import io
import os

# Initialize the SaT model
sat = SaT("sat-3l-sm")
sat.half().to("cuda")

@spaces.GPU(duration=59)
def segment_text(input_text, zip_file):
    results = {}
    
    if input_text:
        # Process single text input
        sentences = sat.split(input_text)
        results["input_text"] = {"segments": sentences}
    elif zip_file is not None:
        # Process zip file
        with zipfile.ZipFile(zip_file.name, 'r') as zip_ref:
            for file_name in zip_ref.namelist():
                if file_name.endswith('.txt'):
                    with zip_ref.open(file_name) as file:
                        content = file.read().decode('utf-8')
                        sentences = sat.split(content)
                        results[file_name] = {"segments": sentences}
    
    # Create a JSON object with the results
    json_output = json.dumps(results, indent=2)
    
    return json_output

# Create the Gradio interface
iface = gr.Interface(
    fn=segment_text,
    inputs=[
        gr.Textbox(lines=5, label="Input Text (Optional)"),
        gr.File(label="Upload ZIP file (Optional)", file_types=[".zip"])
    ],
    outputs=gr.JSON(label="Segmented Text (JSON)"),
    title="Text Segmentation with SaT",
    description="This app uses the SaT (Segment any Text) model to split input text into sentences and return the result as JSON. You can input text directly or upload a ZIP file containing multiple .txt files. All credits to the respective author(s). Github: https://github.com/segment-any-text/wtpsplit/tree/main",
    examples=[
        ["This is a test This is another test.", None],
        ["Hello this is a test But this is different now Now the next one starts looool", None],
        ["The quick brown fox jumps over the lazy dog It was the best of times, it was the worst of times", None],
    ]
)

# Launch the app
iface.launch()