After rolling out video chapters to all creators back in May, which enables creators to define specific segments of their video via time stamps, YouTube is now running an experiment which will see chapters automatically inserted into selected videos, using machine learning to identify segments within each clip.
As explained by YouTube:
“We want to make it easier for people to navigate videos with video chapters, so we are experimenting with automatically adding video chapters (so creators don’t have to manually add timestamps). We’ll use machine learning to recognize text in order to auto-generate video chapters. We’re testing this out with a small group of videos.”
That could eventually see chapters added to all YouTube clips, without creators having to manually add them, giving YouTube more ways to define specific segments related to more search queries.
As noted, YouTube added video chapters back in May, which creates specific segments on the playback bar to signify each part of the clip.
That makes it easier for viewers to skip through to the sections they’re most interested in, but it also enables YouTube, and by extension Google, to identify very specific video elements in response to user queries.
So if you go looking for ‘videos about clipping your dog’s nails’, Google could now provide you with not only a listing of video clips but links to the specific segments within each video that relates directly to your query.
Given this, the addition of automated segmentation could significantly improve YouTube’s search capacity, splicing each video into its key elements – but then again, it could also end up framing segments into weird spaces, and ruining the flow of your content.
YouTube says that creators can opt-out of the experiment, though it will only be trialed on a small scale at this stage.
But it could mean that you see a lot more segmentation and chapters in YouTube clips soon, which has direct implications for search.