app

app.py

@@ -1,6 +1,20 @@
 import gradio as gr
 from PIL import Image
 import pytesseract
+from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
+
+model_name = "google/flan-t5-base"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
+
+def process_text(text)
+    text = text
+    prompt = f"Extract name, dob, license_number, expiry, address from the following text and return JSON:\n{text}"
+
+    inputs = tokenizer(prompt, return_tensors="pt", max_length=1024, truncation=True)
+    outputs = model.generate(**inputs, max_length=256)
+    json_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    print(json_text)
 
 # Function to process image and extract text
 def extract_text_from_image(image):
@@ -29,7 +43,7 @@ def extract_text_from_image(image):
         """
 
         # Return results
-        return extracted_text, img, status
+        return process_text(extracted_text), img, status
 
     except Exception as e:
         error_status = f"""