Update app.py

app.py

@@ -5,7 +5,7 @@ import shutil
 from PIL import Image
 from transformers import ViltProcessor, ViltForQuestionAnswering, pipeline
 from gtts import gTTS
-import pytesseract
+import easyocr
 import torch
 import tempfile
 import gradio as gr
@@ -19,6 +19,9 @@ vqa_model = ViltForQuestionAnswering.from_pretrained("dandelin/vilt-b32-finetune
 # Load image captioning model
 captioner = pipeline("image-to-text", model="nlpconnect/vit-gpt2-image-captioning")
 
+# Load EasyOCR reader
+reader = easyocr.Reader(['en'])
+
 def classify_question(question: str):
     question_lower = question.lower()
     if any(word in question_lower for word in ["text", "say", "written", "read"]):
@@ -36,7 +39,8 @@ def answer_question_from_image(image, question):
 
     if mode == "ocr":
         try:
-            text = pytesseract.image_to_string(image)
+            result = reader.readtext(image)
+            text = " ".join([entry[1] for entry in result])
             answer = text.strip() or "No readable text found."
         except Exception as e:
             answer = f"OCR Error: {e}"
@@ -89,4 +93,4 @@ app = gr.mount_gradio_app(app, gui, path="/")
 
 @app.get("/")
 def home():
-    return RedirectResponse(url="/")
+    return RedirectResponse(url="/")