feat: Add gradio_pdf library for PDF handling and processing

app.py

@@ -9,6 +9,7 @@ from pathlib import Path
 from pdf2image import convert_from_path
 import spaces
 from threading import Thread
+from gradio_pdf import PDF
 
 models_supported = {
     "arabic-small-nougat": [
@@ -152,7 +153,7 @@ with gr.Blocks(title="Arabic Nougat") as demo:
     with gr.Tab("Extract Text from PDF"):
         with gr.Row():
             with gr.Column():
-                pdf = gr.File(label="Input PDF", type="filepath")
+                pdf = PDF(label="Input PDF")
                 model_dropdown = gr.Dropdown(
                     label="Model", choices=list(models_supported.keys()), value=None
                 )

requirements.txt

@@ -5,4 +5,5 @@ spaces
 python-Levenshtein
 pillow
 pathlib
-nltk
+nltk
+gradio_pdf