AI looks at what gets you pumped in video storytelling

Credit: CC0 Public Domain

(Tech Xplore)—"Wow. Pass me a tissue. That was a great story. Oscar worthy. Who wrote the script? What?"

If the answer was a machine then it would be a tall story. Last year, there was a bot -written screenplay, said an article from McKinsey. We wouldn't worry about computers winning Oscars any time soon.

As reported in McKinsey: One character in the film, Sunspring, coughed up an eyeball. A critic said the dialogue often sounded like a series of unrelated sentences.

If machines weren't yet up to top snuff in replacing humans as script writers, the authors of the McKinsey article asked, what about another scenario, asking, "could machines work alongside humans to improve the storytelling process?"

That answer may be worth considering, as in how technology could "supercharge" storytellers, as the authors put it.

Enter researchers from the Massachusetts Institute of Technology (MIT). It has investigated the potential for such machine–human collaboration in video storytelling. The team asked if machines could identify common emotional arcs in video stories. Think about it.

Working as a collaboration between MIT's Lab for Social Machines and McKinsey's Consumer Tech and Media team, they came up with machine-learning models that rely on to watch small slices of video—"movies, TV, and short online features—and estimate their positive or by the second."

They looked beyond plot line to aspects including the score and close-ups of a person's face. "Using these clues, the project's machine learning algorithms were able to identify positive and negative emotions, and map out the extend to which each scene would provoke emotional responses—something the researchers called "visual valence," said Variety.

Why? "When the content of each slice is considered in total, the 's emotional arc emerges."

If AI is useful at this level of analysis, one might consider its taking on a role as script buddy.

Video storytellers may as a result want to make edits to increase engagement; "That could mean a new musical score or a different image at crucial moments, as well as tweaks to plot, dialogue, and characters."

To be sure, we might be looking at a new way in which video stories are created. The script writer might check with machines as part of the writing process.

So what do we have here?

Irish Examiner: "That could mean a new musical score or a different image at crucial moments, as well as tweaks to plot, dialogue, and characters."

McKinsey: As these tools become more readily available, we could see a major change in the way stories are created. Writers and storyboarders "might work alongside , using AI capabilities to sharpen stories and amplify the emotional pull."

More information: www.mckinsey.com/industries/me … s/ai-in-storytelling

47 shares