• capital@lemmy.world
    link
    fedilink
    English
    arrow-up
    24
    ·
    7 hours ago

    Seeing as these ads will be targeted and of varying length, I wonder if a SponsorBlock-like extension with the ability to accept training data from users to help identify ads.

    The Plex server application has a feature which scrubs videos and identifies intros so you can skip them like you can on Netflix. Wouldn’t it be sort of like that?

    Seems like a good use of AI/ML.

    • helenslunch@feddit.nl
      link
      fedilink
      English
      arrow-up
      1
      ·
      2 hours ago

      I wonder if a SponsorBlock-like extension with the ability to accept training data from users to help identify ads.

      That’s…how SponsorBlock works? The ads come at different entry and exit points for every user. Otherwise it wouldn’t be a problem for sponsorblock.

      The Plex server application has a feature which scrubs videos and identifies intros

      Pretty sure they just use timestamps from a crowdsourced database, just like sponsorblock.

      • overcast5348@lemmy.world
        link
        fedilink
        English
        arrow-up
        1
        arrow-down
        1
        ·
        47 minutes ago

        The ads come at different entry and exit points for every user.

        They’re not referring to the YouTube ads, but the “let’s take a minute to talk about today’s sponsor nordvpn that I used on my trip to Antarctica.” This is a part of the video file itself, and it starts and ends at the same time for all users.

        • helenslunch@feddit.nl
          link
          fedilink
          English
          arrow-up
          1
          ·
          edit-2
          27 minutes ago

          This is a part of the video file itself, and it starts and ends at the same time for all users.

          Except it doesn’t when a YouTube ad is injected in the middle. Then all timestamps after the ad are offset by the length of the ad. That’s not from me, that’s from SponsorBlock themselves in the OP.