Update app.py

app.py

@@ -7,14 +7,14 @@ from transformers import pipeline
 from gtts import gTTS
 import tempfile
 import os
-import easyocr
+import pytesseract  # ✅ Replacing easyocr
 
 app = FastAPI()
 
 # Models
 caption_model = pipeline("image-to-text", model="nlpconnect/vit-gpt2-image-captioning")
 vqa_model = pipeline("visual-question-answering", model="dandelin/vilt-b32-finetuned-vqa")
-ocr_reader = easyocr.Reader(['en'], gpu=False)
+
 
 def process_image_question(image: Image.Image, question: str):
     if image is None:
@@ -24,9 +24,8 @@ def process_image_question(image: Image.Image, question: str):
         # Convert image to numpy
         np_image = np.array(image)
 
-        # OCR text
-        ocr_texts = ocr_reader.readtext(np_image, detail=0)
-        extracted_text = "\n".join(ocr_texts)
+        # OCR text using pytesseract
+        extracted_text = pytesseract.image_to_string(image)
 
         # Caption
         caption = caption_model(image)[0]['generated_text']