1

I am new to ML/DL and looking for a good way to generate a handwritten (simulated) file given 2 inputs:

  1. A set of sample handwritten notes (for training). All notes will be from the same person.
  2. A text file.

The output will be a text file in handwritten form, making use of the sample set.

Any pointers would be appreciated. I looked at this, but it differs from my requirements in that it does not generate for a given sample set but an arbitrary form.

The Pointer
  • 157
  • 6
Sriram
  • 11
  • 2
  • there is an obvious and a more involved way. The obvious way is to extract from the notes the letters needed and combine them to form other sentences and texts. The involved approach is to train, for example, a [VAE](https://towardsdatascience.com/understanding-variational-autoencoders-vaes-f70510919f73) – Nikos M. May 13 '21 at 15:36

1 Answers1

0

You need to create a generative model to do this. Conditional GAN will allow you to do that. You supply the sample to condition the output in that way. But, you will need paired data for that. So, given you have a text file, you have to generate handwritten samples for them (use this).

Finally, you have your paired data for cGAN and the sample you want to condition the generation on. All the best.

Abhishek Verma
  • 672
  • 3
  • 4