I would like to generate Text based on Keywords. This means, not specific seed text for a Language Model to complete. Rather a neural network, which given some keywords, would generate a text of x words long, including as most of these keywords as possible.
For example, for a domain specific network in Fashion, with the Keywords
light, dress, orange, cotton. It could output:
This gorgeous orange summer dress is great for wearing on sunny camping days. It's cotton fabric makes it very comfortable to wear.
I’ve seen people are trying GANs to do this. I was wondering if any has experience with this, implemented this, or would recommend to try it in a different way (other than GANs)?
Any advice or hint is mostly welcome!