Keep text and back to STORY LIVE


This creative support tool (CST) was devised and the site programmed by Geoff Davis for post-graduate research at UAL CCI 2020. This is a simplified version which sends any text created to the user by email. Or it can just be used, and copied from.

Geoff Davis also produced Story Software zooming image/text noteboard Notes Story Board v 2.2

Geoff Davis Organisation website

Text Synth

Geoff Davis:
A publicly available text generator was used here, as this is the sort of system people might use outside of the study.
It was also not practical to recreate (program, train, fine-tune, host) a large scale text generation system for this usability pre-study.

Fabrice Bellard, coder of Text Synth:
Text Synth is build using the GPT-2 language model released by Google OpenAI. It is a neural network of 1.5 billion parameters based on the Transformer architecture.

GPT-2 was trained to predict the next word on a large database of 40 GB of internet texts. Thanks to myriad web writers for the training data and OpenAI for providing their GPT-2 model.

For details on how to change the parameters, should you wish to, please see Text Synth Technical Notes from the experiment's left panel.

Permission was granted to use Text Synth in the study by Fabrice Bellard Jul 7 2020.
Powered by textsynth.

Visit OpenAI's blog for more inforamtion on Google's OpenAI text generation.

Test dog image

The image is from public sources under free license.


Who owns the generated text?

This summary is from Adam King (coder of Talk to Transformer):

This seems to be a complicated issue. We can't give legal advice, so if you rely a solid answer you'll need to consult a legal professional.

We do however waive any rights we may have in the text you generate and grant you licence to use it for any purpose (to the extent that we have that right), royalty-free and without warranty. You don't have to credit us anywhere.

To hint at the complexity of this issue, consider that the neural network is:

1) designed and trained by OpenAI who licensed its source under the MIT license,

2) learned from millions of web pages containing content in which many people hold copyright,

3) is hosted by us (we waive all rights to the content) and

4) is conditioned on the prompt you give it (your own content).