Realistic Handwriting Generation Using Recurrent Neural Networks and Long Short-Term Networks

  • Suraj Bodapati
  • Sneha ReddyEmail author
  • Sugamya Katta
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 1090)


Generating human-like handwriting by machine from an input text given by the user may seem as an easy task but is very complex in reality. It might not be possible for every human being to write in perfect cursive handwriting because each letter in cursive gets shaped differently depending on what letters surround it, and everyone has a different style of writing. This makes it very difficult to mimic a person’s cursive style handwriting with the help of a machine or even by hand for a matter of fact. This is why signing names in cursive is preferable on any legal documents. In this paper, we will try to use various deep learning methods to generate human-like handwriting. Algorithms using neural networks enable us to achieve this task, and hence, recurrent neural networks (RNN) have been utilized with the aim of generating human-like handwriting. We will discuss the generation of realistic handwriting from the IAM Handwriting Database and check the accuracy of our own implementation. This feat can be achieved by using a special kind of recurrent neural network (RNN), the Long Short-Term Memory networks (LSTM).


Handwriting generation Recurrent neural networks (RNN) Long Short-Term Memory networks (LSTM) IAM handwriting database 


  1. 1.
    Getting Started-DeepLearning 0.1 documentation., 2017 [Online]. Available Accessed 19 Feb 2018.
  2. 2.
    Bridle, J.S. 1990. Probabilistic Interpretation of Feedforward Classification Network Outputs, with Relationships to Statistical Pattern Recognition. In Neurocomputing: Algorithms, Architectures and Applications, ed. F. Fogleman-Soulie, and J. Herault, 227–236. Springer.Google Scholar
  3. 3.
    Perwej, Y., and A. Chaturvedi. 2011. Machine Recognition of Hand Written Characters Using Neural Networks. International Journal of Computer Applications 14 (2), 6–9.Google Scholar
  4. 4.
    LeCun, Y., L. Jackel, P. Simard, L. Bottou, C. Cortes, V. Vapnik, J.S. Denker, H. Drucker, I. Guyon, U. Muller and E. Sackinger. 2018. Learning Algorithms for Classification: A Comparison on Handwritten Digit Recognition. Neural Networks: The Statistical Mechanics Perspective 261 (1), 276–309.Google Scholar
  5. 5.
    Plamondon, R., and S. Srihari. 2000. On-Line and Off-Line Handwriting Recognition: A Comprehensive Survey. IEEE Transactions on Pattern Analysis and Machine Intelligence 22 (1): 63–78.CrossRefGoogle Scholar
  6. 6.
  7. 7.
    Leverington, D. Neural Network Basics,, 2009. [Online]. Available Accessed 19 Feb 2018.

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  1. 1.Department of Information TechnologyChaitanya Bharathi Institute of TechnologyHyderabadIndia

Personalised recommendations