17 December 2020

Creating new story formats from text articles - replay our webinar

Can we quickly create comics, summaries, image galleries and even videos using just the text of online news stories? At News Labs we are very excited about the potential the latest technologies offer for the semi-automated creation of these other story formats.

We've been running experiments throughout this year to help answer this question and hopefully reach all audiences with news output in the formats that best suit them.

We were delighted when the team from the JournalismAI project at Polis, the journalism think-tank at the London School of Economics, invited us to share this latest work at their week-long festival.

The webinar, including questions and answers, is embedded below and can also be viewed on the event webpage.

In it our executive product manager, David Caswell, outlines the challenges facing the news industry. Audiences now have near infinite choice in an online world, instead of just a few broadcasters and publications having the resources to reach audiences. Can semi-automation help public service news providers offer the multiple formats and personalisation that consumers are growing accustomed to?

We share the details of the Graphical Storytelling prototype, which automates the creation of mobile compatible comics from text stories. It is now being trialled by our social media and health teams to create visual stories on Instagram and Facebook.

Our latest project that follows on from this work is Modus. In it we explored automated summarisation and entity extraction to create bullet points, captions and still image galleries. Senior journalist Conor Molumby discusses the early findings from that project and David also talks about how this could feed into our work on personalisation in 2021.


Love data and code?

We're looking for talented developers to join our team.