YouTube normalizes to negative 14 LUFS, Spotify to negative 14, podcasts often negative 16 to 19. What target does the panel actually hit and do you master per platform?

7:22This is ridiculous. She goes, oh no, I really like that thing. She was also notorious for buying a lot of stuff off QVC. Anyway, Alexander, let's go to the next question. All right. This next question, by the way, that previous question was submitted by Showhub or in Showhub. This is Paul Vahos in Austin, Texas. YouTube normalizes to negative 14 luffs, Spotify to negative 14, podcasts often negative 16 to 19.

7:48What target does the panel actually hit and do you master per platform? Let's talk about this. Ronnie, you go first. Yeah. So there are a few aspects of this. So, of course, every platform has some kind of most fitting number. As you said, YouTube is minus 14. I would say what is more important is that you master your loves or master output to the genre you are doing.

8:22So if this is a podcast, of course, you could have more dynamics in it, thus having a bigger number. So more minus. Here in Office Hours, we tend to hit minus 24. And the reason for that is we are a kind of a high quality podcast where we always can have enough headroom so that Chris Fenwick can have his outbursts. That guy is intolerable.

8:49And in YouTube, you normally have less headroom or less kind of dynamics. And of course, if you listen to EDM music, which I'm kind of known of doing sometimes in the weekends, there's even less dynamics, typically minus six or up to minus three dB LUFS. And the number is is very complex and I'm pretty sure some of my co-panelists will take some time and explain

9:21the inner details of that, but that is the general idea. And of course, as an answer to your question, yes, I am mastering to the different platforms needs and leave it up to me to decide the dynamics, not the platform to re-encode or do stuff with my signal. So I try to optimize to what they ask for. for. Alexander? There's a lot of subjectivity with this stuff because yes, you're right. First of all,

9:51I just want to point out that podcasts don't actually have a ratified standard. So there are, there is a standards body that determines loudness, recommended integrated program loudness, depending on whether or not you're in the European Union or in the United States or wherever you are in the world there are different standards so negative 16 has kind of been an unofficial agreed upon target for distribution of stereo audio for podcasts and negative 19 for mono

10:23because there's the center pan law so there's about a 3db difference there so i also wanted to point out too that these platforms don't actually reprocess your audio I believe there's last time I looked at Apple Music it doesn't do there's something called sound check in Apple Music Which will loudness normalize, but you have to turn it on and I don't know if Spotify does that

10:44But I'm pretty sure these platforms don't reprocess it because I've actually Had a look at music that was distributed on these platforms And I've for example like you typically modern pop rock and metal music is Mixed very very loudly and I've seen music on these platforms hit negative 9 l UFS Like I've actually measured the audio, the overall program audio, and I've seen it be very, very loud.

11:08So I don't think they actually reprocess that audio unless they've changed that in the last couple of years. It is optional. It is a user preference. You can turn on in Spotify and Apple Music to loudness normalize. So you need to decide what you want to do. Now, there's obviously a lot of bad audio that is distributed on these platforms, too. I've measured stuff where it's not just the so the LUFS measurement is one thing

11:32Paul which is typically we're talking integrated loudness was it which is a measurement over the entire program material because that gives you a better idea perceptually how human hearing is but there's also other factors too like like are you actually measuring inter sample peaks so sometimes I've seen audio that meets the loudness target but then but then they actually have clip clipped audio above zero dBFS. So that's also a problem. So you also need to really,