diff --git a/newspaper-extractor/backend/app/pipeline/ocr_engine.py b/newspaper-extractor/backend/app/pipeline/ocr_engine.py
index 66bab99..1418aa3 100644
--- a/newspaper-extractor/backend/app/pipeline/ocr_engine.py
+++ b/newspaper-extractor/backend/app/pipeline/ocr_engine.py
@@ -36,4 +36,5 @@ class OCREngine:
 
     def detect_language(self, crop):
         # Quick langdetect on a fast English OCR pass, or script detection
-        return "en"  # plug in script-detection model
+        # Hardcoding to 'te' (Telugu) which also supports English numbers/characters
+        return "te"