Any builin Library for Data synthesis for OCR

Hello the expert community,

I am interested to train OCR for hand written numerals. I have taken 400 images of hand written. Yet I didn't apply any challenging transformation such as rotation, blurring, lighting etc. Now I am interested to synthesis the data to increase the number of training instances and also generate data. Is there any built-in library for this? And what kind of transformation are usually applied for OCR synthesis Dataset?