LSTM based Bangla OCR engine Omar Uncategorized Comments It is a CNN+LSTM+CTC based Bangla OCR engine and trained with 18000 words in only one font. It is written in C# 2012 without any external dependency.
