▲ 987 ▼ YouTube is experimenting with server-side ads (lemmy.world) submitted 2 years ago* (last edited 2 years ago) by Dju@lemmy.world to c/enshittification@lemmy.world 352 comments fedilink hide all child comments https://fosstodon.org/@sponsorblock/112603139898164385
[–] Matty_r@programming.dev 15 points 2 years ago (1 child) We could use audio fingerprinting to detect ads in the buffer permalink fedilink source parent hideshow 2 child comments replies: [–] zer0squar3d@lemmy.dbzer0.com 11 points 2 years ago (1 child) Yeah that could work... What about creating some sort of *arr for YouTube videos that downloads them and processes them with some sort of AI audio/video processing to remove the ads and recombine the video. Youtubarr it could be called. If we really want we can also remove the ads from the creator in the video too. It would still count as a view to the video too so creator won't lose out on money. Anyone with objections to this? permalink fedilink source parent hideshow 2 child comments replies: [–] Cubes@lemm.ee 3 points 2 years ago It's a neat idea, but computer vision stuff can get quite computationally expensive when done locally and is prone to input poisoning attacks (especially if the models used are open source). Not saying it wouldn't be possible, but I think some of the other ideas posed here would be better starting places. permalink fedilink source parent
[–] zer0squar3d@lemmy.dbzer0.com 11 points 2 years ago (1 child) Yeah that could work... What about creating some sort of *arr for YouTube videos that downloads them and processes them with some sort of AI audio/video processing to remove the ads and recombine the video. Youtubarr it could be called. If we really want we can also remove the ads from the creator in the video too. It would still count as a view to the video too so creator won't lose out on money. Anyone with objections to this? permalink fedilink source parent hideshow 2 child comments replies: [–] Cubes@lemm.ee 3 points 2 years ago It's a neat idea, but computer vision stuff can get quite computationally expensive when done locally and is prone to input poisoning attacks (especially if the models used are open source). Not saying it wouldn't be possible, but I think some of the other ideas posed here would be better starting places. permalink fedilink source parent
[–] Cubes@lemm.ee 3 points 2 years ago It's a neat idea, but computer vision stuff can get quite computationally expensive when done locally and is prone to input poisoning attacks (especially if the models used are open source). Not saying it wouldn't be possible, but I think some of the other ideas posed here would be better starting places. permalink fedilink source parent
