Create app.py

app.py

@@ -0,0 +1,31 @@
+import gradio as gr
+from pdf2image import convert_from_path
+from PIL import Image
+import pytesseract
+import os
+
+# Set Gujarati as OCR language
+OCR_LANG = "guj"
+
+def extract_gujarati_text(pdf_file, page_number):
+    # Convert selected page to image
+    images = convert_from_path(pdf_file.name, first_page=page_number, last_page=page_number)
+    image = images[0]
+    image_path = f"/tmp/page_{page_number}.png"
+    image.save(image_path)
+
+    # Run OCR with Gujarati
+    text = pytesseract.image_to_string(Image.open(image_path), lang=OCR_LANG)
+    return text
+
+with gr.Blocks() as demo:
+    gr.Markdown("## 📚 Gujarati OCR from PDF (Tesseract-powered)")
+    
+    pdf = gr.File(label="📤 Upload Gujarati PDF", file_types=[".pdf"])
+    page = gr.Number(label="📄 Page Number", minimum=1, value=1, step=1)
+    button = gr.Button("🔍 Extract Text")
+    output = gr.Textbox(label="📝 Extracted Gujarati Text", lines=20)
+
+    button.click(fn=extract_gujarati_text, inputs=[pdf, page], outputs=output)
+
+demo.launch()