this post was submitted on 05 Apr 2024

141 points (98.0% liked)

Asklemmy

43897 readers

943 users here now

A loosely moderated place to ask open-ended questions

Search asklemmy 🔍

If your post meets the following criteria, it's welcome here!

Open-ended question
Not offensive: at this point, we do not have the bandwidth to moderate overtly political discussions. Assume best intent and be excellent to each other.
Not regarding using or support for Lemmy: context, see the list of support communities and tools for finding communities below
Not ad nauseam inducing: please make sure it is a question that would be new to most members
An actual topic of discussion

Looking for support?

Looking for a community?

Lemmyverse: community search
sub.rehab: maps old subreddits to fediverse options, marks official as such
[email protected]: a community for finding communities

~Icon~ ~by~ ~@Double_[email protected]~

founded 5 years ago

MODERATORS

[email protected]

141

Are there any legal issues recreating YouTube SponsorBlock for Podcasts? (lemmy.ml)

submitted 7 months ago by [email protected] to c/[email protected]

56 comments fedilink hide all child comments

I'm part of a small group of Jr Self Taught Web Developers who were recently brainstorming ideas for a Group Project App we could put together and actually create a user base.

I offered up the suggestion of a podcast application which would have the major feature of being akin to YouTube Sponsor Block, but specifically for podcast episodes.

Essentially, a user contributed database of timestamps for podcast episodes where the mention of cutting to sponsored ads or mentions of sponsorships would be marked so they could be edited out of the episode and then the user could also download said episode where ads are cut out of the final audio file.

My idea was shot down due to fears of possibly infringing on copyright and we ended up with going with another idea. I'm certainly not upset, and am actually excited with the project idea we did choose, but it did get me wondering about whether this idea actually could have legal implications.

I know specifically with YouTube there appears to be a sort of legal loophole that prevents Google from suing projects like invidious, yt-dlp, and YouTube Sponsor Block, but am unaware of the specific details as to how this works.

Thusly, I just wanted to ask if anyone has any insights into whether this project idea would incur any legal infractions from the likes of IheartRadio and other media platforms?

To be clear, I'm not seeking legal advice here, and I'll be taking any responses with a grain of salt, but I just wanted to see if anyone knows anything on this subject and the legal concerns raised.

I very much dislike being advertised to and podcasts are one of the last bastions of media where advertisements still come up regularly and I'd love to make this application for those who are frustrated with how often they have to skip through sponsor mentions.

Thanks in advance.

top 50 comments

sorted by: hot top controversial new old

[–] [email protected] 2 points 6 months ago (1 children)

Take multiple downloads of a podcast from different locations, automatically filter out any audio that is unique to either stream. This can be done with hashing or other means. This would address dynamic advertisements.

Everything left over can be manually time stamped by volunteers to bracket the ads.

Alternatively. Take multiple podcasts from a series and filter out all audio they have in common. This would handle static ads, intro music, etc.

[–] [email protected] 1 points 6 months ago

Nice! That sounds like a solid approach. Right now I'm just solidifying a strategy before possibly consulting with a digital rights lawyer. This is probably a few years in the making, but I'm gathering ideas from various communities and all insights are helpful!

Thank you!!

[–] [email protected] 4 points 7 months ago (1 children)

Other answer seems to suggest that the problem is that the same podcast can be available, depending on where and who is listening to it, with different length due to different ads injected into. Here's my probably stupid and completely ignorant suggestion: instead of using timestamps for both begin and end of the ads segment, you could use a timestamp for the beginning, and an hash of the first part of "non-ads" segment. I'll try to explain better:

|----------------xxxxx--------------------|
                ^     |___|

The xxx is the ads segment, the ^ is the timestamp of the beginning of the ads, the |___| is a small duration segment (for example, 0.5 seconds) right after the ads segment. The data of that segment is hashed and used as "end ads segment indicator".

On the other device, with a different duration of the ads, you should start hashing it to find the corresponding segment.

Is this doable or did I just said a bunch of idiot things?

[–] [email protected] 2 points 7 months ago

Possibly doable, and definitely not a bunch of idiot things.

The beginning of the FIRST ad certainly does start at a consistent point during the podcast episode, but due to the dynamic nature of the injected targeted ads afterwards, the remaining timestamps for the beginning of the subsequent ads would be different. The hash of the audio file was suggested by another helpful person earlier, and it has piqued my interest, though it's implementation, at least as I conceive of it currently, would be rather clunky, as it would require an ad free version of the audio file to compare it against, as well as a way (possibly a hash or some sort of audio recognition software, or AI trained to recognize ads, effectively acting as audio recognition software), to recognize exactly when the ads started and stopped by comparing the ad free version second by second to the ad-injected version.

Additionally, because re-distribution of said audio files would definitely land me in legal trouble, I'd have to dynamically generate those timestamps and send it to you while you received the audio file from the official distributor, all to be edited on your device upon arrival.

I'm still very much a noob at programming and webdev, so this is definitely something that is probably a few years down the line in the making as I continue to upskill, but it's good to think about as I would like to produce this if at all feasible (and won't land me in legal trouble or hurdles, I'd like this to be akin to Invidious and Sponsorblock if possible).

Thanks for the suggestion! You definitely got me thinking hard on this problem and its potential solutions.

[–] [email protected] 5 points 7 months ago

Seems like time-shifting (VCRs, Tivo) is protected, but I agree with advice to consult a lawyer.

[–] [email protected] 23 points 7 months ago

Retransmission of a podcast from your own server - no.

Cutting sponsored fragments on the end device - yes.

At least in most countries.

[–] [email protected] 15 points 7 months ago* (last edited 7 months ago) (1 children)

edited out of the episode and then the user could also download said episode where ads are cut out of the final audio file

This is your problem, because you're redistributing someone else's work with the ads cut out, which isn't sufficiently transformative to qualify for fair use. Sponsorblock is allowed because it doesn't actually interfere with the video stream, it just tells your computer when to skip ahead using YouTube's already-existing playback features - your app should work the same way, integrating into an existing podcast platform and skipping forward based on crowdsourced timestamps, then the only thing you're providing are the timestamps, which don't violate copyright.

[–] [email protected] 7 points 7 months ago

Exactly this. What you'd want is to develop your own podcatcher where instead of the listener pressing skip to skip forward 1 minute manually until they get to the end of the ad, your app does it automatically.

[–] [email protected] 9 points 7 months ago

I know nothing of the implications of developing what you’re proposing, but I 1000% support it and would actually start listening to podcasts again if it ever came to fruition and I could use it.

[–] [email protected] -2 points 7 months ago (2 children)

Am I wrong in thinking that if you obscure your identity well enough, there's no reason to worry about legality?

[–] [email protected] 1 points 7 months ago

I thought about this, and I believe I just wouldn't want to increase my threat profile that much. I'd have to put the API up on either a .onion or i2p vps instance, in which case I'd have to brush up my knowledge on the dark web more than I currently have. I'd also have to become a much better cybersecurity expert than I am now. Don't get me wrong, these are all great skills to have, but if I can go the more legit route of being able to do this legally and without fear of unnecessary legal hurdles (frivolous lawsuits, etc.), then I'd prefer that.

[–] [email protected] 8 points 7 months ago* (last edited 7 months ago)

I mean, only insofar as that's true of anything illegal?

[–] [email protected] 7 points 7 months ago (2 children)

As much as I hate ads on podcasts that still wouldn't be a good enough reason for me to switch apps unless the new one is atleast as good in other ways too as the one I'm using now.

[–] [email protected] 0 points 7 months ago

NewPipe has variations that include additional features like incorporating Youtube SponsporBlock API. I had thought on forking an existing podcast application and simply adding this API once I figure out how to generate the timestamps.

[–] [email protected] 2 points 7 months ago

Ive seen sponsor block for YTb added to a few different playes. It sesms it operates moren like a DB that other clients can hook into. So OPs idea wouldn't need to be a new podcast player exactly.

[–] [email protected] 25 points 7 months ago (2 children)

From what I understand, they're able to practically make custom audio files for every download. Sharing the time stamps wouldn't work that well. Re-distributing podcasts without the ads would definitely land you in legal trouble, cause every audio file is their "work of art".

Not a problem for ublock because you're editing their work of art for your personal use, and sharing unaltered stuff.

And youtube sponsor block is just sharing time stamps you might be interested in.

AI system that can recognize patterns and auto skip forward?

[–] [email protected] 1 points 7 months ago

Yes, using a trained AI model that recognizes ad segments could be possible for this, albeit expensive due to the cost of GPUs on a VPS.

[–] [email protected] 6 points 7 months ago (1 children)

Maybe if you could distribute audio files (or hashes of audio files) that mark the start and stop of ads, that would solve the problem.

I guess podcasters could combat this by inserting random noise into their audio files, but they probably wouldn't do that.

[–] [email protected] 4 points 7 months ago (2 children)

They don’t need to add random noise, they just do what they already do and insert new advertising materials. Your static timestamps mean the ads and content end up at different locations.

[–] [email protected] 1 points 7 months ago

Yeah, that's a problem. Dynamic length of targeted advertisement breaks would mean even a user generated database of timestamps wouldn't be that useful.

[–] [email protected] 7 points 7 months ago (2 children)

I'm not suggesting static timestamps, but small audio files of the podcast about to enter, and just exited an ad.

The app could then search for the clips in the podcast to get the timestamp.

If there are copyright issues of sharing small clips, you can just save a hash of a clip, which will allow the app to find a match, but is not itself the Intelecual property of the podcaster; The hash cannot be turned back into the audio file. The hash would be smaller than the audio clip anyway, so sharing hashes would be better

[–] [email protected] 1 points 7 months ago

That's an intriguing approach. Saving a hash of the clip instead of the timestamps MIGHT work. I'm still a bit worried about legal ramifications in that case.

[–] [email protected] 6 points 7 months ago (1 children)

Some perceptural hash of the actual ads could work to. You could run into legal trouble sending the ads themselves or the hosts speaking.

[–] [email protected] 2 points 7 months ago (1 children)

Good idea! I bet you could make good ad library by comparing the audio between episodes of the same podcast (to catch the ads read by the host) and between different podcasts (to catch the targeted ads inserted into a lot of podcasts)

[–] [email protected] 1 points 7 months ago

Such a library of ads would quickly become massive, even if stored in a series of hash references. Interesting idea though.

[–] [email protected] 6 points 7 months ago (1 children)

so they could be edited out of the episode and then the user could also download said episode where ads are cut out of the final audio file.

This is the part that might be problematic and I can see being part of a civil suit (I am not a lawyer). Depending on how you collect and store the episodes (which you may not actually have to do to achieve your goal, but is the easiest solution) you would likely run afoul of "distribution" precedents in the US that may result in a judgement against you.

But even if you didn't actually break the law, the media lobbies globally are well known for filing huge numbers of lawsuits over anything that even looks a little like it might be costing them money. Defending yourself at all is hard time-consuming and often expensive. It's not something I would recommend going into casually.

https://torrentfreak.com/category/lawsuits/ is a great site for learning about the current lawsuits from a tech perspective, and has helped me out many times over the last decade. It's one of the gems of the internet, in my opinion.

[–] [email protected] 0 points 7 months ago

I love that site.

Ultimately, as others have suggested, the most probable way of doing this legally would be to not distribute anything other than timestamps as well as a simple binary/executable that would use a built in media editing tool like ffmpeg to cut out the advertisements/sponsor mentions and then recut the audio file back together. This is much akin to how sponsorblock works from what I've gathered so far.

load more comments