Adding Cover Art & Meeting I-85 Compliant Needs

[Dana_Tucker]

I still have a question about this:
EBU R128 recommends to normalize audio at -23 LUFS.
Where does the -16 LUFS for podcasts come from? Has it been published by a US, European or International standards organization? If so, where do I find it?

As far as I know, Thomas Lund, is really the "God Father" of the EUB. (I think). He was in charge of the original P/LOUD Group that started these audio tests. He wrote a paper explaining that his original assumptions of a -23LUFS were insufficient for portable audio devices. This is an excerpt from his findings.

"MOBILE TV AND PODCAST GUIDELINES: A station should think carefully about immediate and future requirements when deciding on the best overall strategy for
the handling of Mobile and Podcast. The procedure needs to be automatic, transparent, well sounding and flexible. Automatic It's a waste of valuable time to prepare content for more than one platform, namely HD. Transcoding to Mobile TV with a Target level between -18 LUFS and -14 LUFS must happen automatically. -16 LUFS is the goal".

This is what Auphonic based their -16LUFS @ a -1DBTP program on. You can read his paper here. http://goo.gl/qti9qr

[Dana_Tucker]

The EBU R-128 is a good thing IMHO.

I've spent a lot of time trying to address the lack of an Audacity plug-in.
The main problem I had (and still have actually) is to adhere to their True Peak measuring system.
It's currently not possible within Nyquist because the resampling algorithm crashes to soon and the convolution is too slow at present.
Fortunately, this will soon be taken care of, the stand-alone Nyquist has already a Fast Convolution algorithm. The new library will certainly soon be integrated in Audacity (not necessarily for 2.1.3, as this will rather be an bug-clean-up version).

Robert

I agree Robert. I have 3 e-mails out to companies that say their meter will conform with the EBU suggestions. The EBU list programs on their site that say they will render your audio to the R-128 specks and two of the meters I have downloaded, are not on their site. All the meters will get me to the -16 LUFS, but not one single one will do it with a -1DBTP, except Audphonic. Now, very seldom do I use something that I can not calibrate with the same usage. I am only assuming that Auphonic is correct.

I carried a $300.00, 6' Macman Level in my truck for 12 years. It only came out when a sub contractor was working for me. I used it to calibrate their level so I did not have to go back and start ripping down walls. They would ask me how I knew my level was the "Good One"? I would simply reply because it cost $280.00 more then yours and Mackman backed it with a $100,000.00, guarantee.

[steve]

So in your opinion Dana_Tucker, would it be correct to say the -16 LUFS is the "de facto standard" for podcasts?

[cyrano]

True. It is strange that this measurement system seems to induce so much resistance. For one part, this seems to stem from people in the live music sector, but also from studio engineers. And every time it is, because people tend to read in a cursory manner. So I understand why the EBU changed the statement.

I am on your side with your above statement and the following statement is my opinion, not a fact. "In the USA, people and myself included, do not want our government to tell us how to listen to our music, or what programs we can watch on TV. And we really don't want them to tell us how loud to play it. (That is why we have laws regarding how loud you can project audio from your car, business or home). In Manassas Virginia, when I reside, the local ordnance has the following equation. "dbs + footage from any exterior wall = citation if exceeded".

I understand the feeling. We're like that too, not liking to be told how to live. But the EBU isn't "the government". And none of EBU128 is more than a recommendation from the tech folk how lawmakers can apply measurement level.

Also true. But then, a podcaster isn't legally obliged to use the EBU compressor. And I don't think many will, as it would provide relatively little benefit. The studio sector seems to understand that it is a great advantage, not a threat.

I must respectfully disagree with your statement. I deal with Podcasters world wide and 99% of them adhere to these standards. The Podcasters that do not adhere to these standards, mostly are the ones just starting out and do not know any better. Once they are informed, they generally conform.

As far as I know the legislation here, the loudness rules in existence only apply to clubs, live music and any public address systems, such as movie theaters. Podacasts, as long as they are not played in public, are private. And private music isn't regulated, as long as you don't disturb the neighbours. That is an area that is outside of EBU128 because of the same problem. Even 45 dB is quite loud if you are trying to sleep.

IANAL and I have no idea about US or other international legislation.

Just trying to avoid OS wars. "Apples" seems to work on some like a red flag, sometimes.

I do not know what "OS wars" mean.

OS = operating system. The Internet wars between fans of Windows, OSX and Linux, mainly.

With "other electronic media", they mean things like recordings, DAB, streaming... Not especially podcasts. When the work was started, the word "podcasting", didn't even exist...

I must again Respectfully disagree. If that is your interpretation, then it is just that, "your interpretation". My interpretation is just what it says. "other electronic media". As far as Podcasting not being around when they started working on their project, originally called the "P/LOUD", the EUB was about 6 years late to the Podcasting Party.

"The term “podcasting” first appeared in an article by Ben Hammersly, the first one to use the term, in an issue of The Guardian. However, Adam Curry is believed to be the first person to actually come up with the idea of putting content into small files and releasing them episodically to the public. It was not until 2004 where the public really began to catch on, and big business started releasing their information over the internet via podcasts, and the public began to take hold of the idea. One of the first companies to make a device specifically used for transporting media such as podcasts was i2Go. They claimed that with their eGo player, in combination with their MyAudio2Go.com website, that users would be able to download episodes of sports, weather, music, and news right to their device, and take it wherever they want. The device also came with software that could automatically download content to the pc and whenever the device was connected, would automatically download to the device. The company then eventually folded". You can read more here, it is a real good article on the history of Podcasting. http://www.ipodder.org/history#ixzz3zTxxakPq

EBU128 started even before the internet. It predates loudness laws and has for a very long time been a discussion between national radio/TV stations, when TV publicity started to affect them. They were annoyed by the very loud publicity destroying their carefully built "soundfield". Just a few weeks ago, the BBC finally decomissioned their original NICAM radio and TV compressors that had been running since the 70's. These got replaced with a home-built system that can be easily updated by changing the software. There simply are no commercial offerings capable of doing what the BBC's new system does. And once again, the BBC has laid out a blueprint that will allow the audio industry to build these new machines.

In the early years of commercial TV, national stations here didn't have publicity. When they were allowed to run publicity, commercial TV and radio stations started their own kind of loudness war: being louder than the rest was the goal. And since publicity was already louder than the shows, shows started getting louder too. That gave an up to 12 dB difference between national and commercial stations. So, the law stated that it should be max. 20%, I believe they copied that from US and Japanese law. But the law doesn't state explicitly how it should be measured. This law meant you could complain if it was really exaggerated, but the main problem still exists.

Technical people from the national networks joined forces to come up with a real loudness measurement for the first time. Of course, years later the commercial stations showed interest too.

BTW there have been a lot of companies trying to do something similar to the eGo. I must have a Sony ebook reader from the early 80's around somewhere, that was to compete with Philips' CDi, but minidisc based. They all failed. I have worked on the launch of several of these devices, but not the eGo, as it never really made it to Europe.

That is part of the problem. Some Americans are clearly incapable of appreciating anything from outside the US of A. You can see the same with any technical system, from the "Imperial" system, down to cell phones...

Well, you nailed that statement Brother! We learned it from the British back in the 1700's and I think a lot of them are still PO'ed about it.

Same here. We still don't trust any Germans

Again, no understanding...The EBU isn't pushing anything. This is a system that was developed on demand and for the European broadcasting sector. But as soon as you try to share knowledge, it seems you're a dangerous commie to some. EBU 128 is a system that probably will solve the loudness war. You can make up your own mind. There's no pressure at all. Use it and benefit from it. No license, not patents, it's free for everyone to enjoy. Or don't use it and risk to be left behind...

I apologize and would like to make the follow correction. When I used the term "pushing the agenda" I meant no disrespect, in fact, I was giving the EUB credit for starting this process. I should have chosen my words more carefully. As far as solving the loudness wars, I think that will happen when all the 1000's of Professional Recording Studios as well as the Basement Shops come together with each other. Your last sentence has me a little confused. "Or don't use it and risk to be left behind".

Brother I preach it, I teach it. It is all over my website. I have done podcast and videos concerning the importance of adhering to it. I have a discount with Auphonic because the EBU has their program listed on their own site. I get close to 30 e-mails a week, asking me to sponsor someones Podcast. I have a 5 questionnaire that you must fill out before I will even respond to you.

Question number 2 asks: "At what LUFS do you render your Stereo or Mono Audio File as well as the Bit Rate?" If you look at the second company listed on my own site, this may clear up some of our confusion. http://www.podcastingschool.org/wp/affiliates/. Rest assured, I am on your side. That is why I started this post. So the thousands of Podcasters that use Audacity as their DAWS, could complete their workflow, from start to finish, using nothing but Audacity.

Sounds like you're a missionary man, to me

And there really is no need to apologize for anything.

I promise you that for myself, if Audacity had this feature when I could get my audio to a -16 or -19 with a -1dbtp, I would never open Reaper again!

Well, KlangFreund's LUFS meter runs in Audacity. I haven't run it through its paces yet, (I'm currently testing plugins) but it seems to work. And it's open source, the code is on github. I could ask the author if he would mind if the Audacity devs borrowed his LUFS code, if needed.

[Dana_Tucker]

So in your opinion Dana_Tucker, would it be correct to say the -16 LUFS is the "de facto standard" for podcasts?

Steve, it is not my opinion, this standard is PREACHED world wide by the top podcasting experts on the face of this earth! (Just so you know, I do not consider myself to be in that group). If you really want to get a $100,000 audio education for free, check out Paul. http://www.producenewmedia.com/. He is very knowledgeable like you are. Sometimes I just want to beat him to death with a dictionary! Half the stuff he talks about I have to look up on Google, just to get the meaning of the word!

[Dana_Tucker]

Well, KlangFreund's LUFS meter runs in Audacity. I haven't run it through it's paces yet, (I'm currently testing plugins) but it seems to work. And it's open source, the code is on github. I could ask the author if he would mind if the Audacity devs borrowed his LUFS code, if needed.

I have an e-mail from the company. They sent it today. The guy was very informative. He is going to make me a video to show me, how to do it. My only request was that it be a -16LUFS @ a -1dbtp. According to him, it can only be done using their meter, along with a peak limiter. I spent 5 hours today, doing that very same set up, no success.

"Hi Dana, thank you for your interest in the LUFS Meter. It depends on your source material, but for most podcasts you will need a limiter together with the LUFS Meter to achieve -16 LUFS and a true peak of less then -1dB(TP).

After I received your mail I tried a few limiters I own and I must admit that FabFilters Pro-L is just incredible good for this job - and for limiting in general. But it comes with a high price tag. I have seen on your website that you're using Reaper. That's why I listened to the available limiters in Reaper and I found the MGA Limiter quite usable for this task.

I try to find time tomorrow to create a video to describe in detail how to achieve this with the LUFS Meter, Reaper and its included MGA Limiter. Btw. a new demo version of the LUFS Meter will be available soon, where you will be able to also test the full LUFS Meter with true peak. Kind regards, Samuel".

I did not have the heart to tell him that most new Podcasters do not have the knowledge to do this. So, for his kindness as well as his effort, I will purchase his product, even-though, I already have a program that is just drag and drop and POOF, all is right with the world again and out comes a -16 with a -1.

[steve]

So in your opinion Dana_Tucker, would it be correct to say the -16 LUFS is the "de facto standard" for podcasts?

Steve, it is not my opinion, this standard is PREACHED world wide by the top podcasting experts on the face of this earth!

What I meant was that it is a "de facto standard" rather than a "de jure standard" (https://en.wikipedia.org/wiki/De_facto_standard).

[Dana_Tucker]

What I meant was that it is a "de facto standard" rather than a "de jure standard" (https://en.wikipedia.org/wiki/De_facto_standard).

Thanks for the reply Steve. Both my Interns are off this week and two are at a out of state conference. I am running short staffed. You can find your answer here out of 78,400 references. My short answer is, all of the above! https://goo.gl/EKaPcF. I think you could judge which standard it should be labeled under, better then I.

I was able to apply a chain effect that will do, what I want to, concerning this issue. One part Audacity, one part from an outside source. I did a 45 min interview over the phone, ran it through the chain, POOF, out came a mono file at a -19LUFS, with a dead on -1DBTP. I will let you know when the details are hashed out.

On a side note. I have also added a chain for Audacity that includes a Normalizer, 11 Band EQ, Soft Peak Limiter, De-esser and a Wav Exporter that puts me dead center of the ACX Requirements. I am aware that the ACX requires a mp3, I think it is best to save the original file in a wav format. Run it through the chain then save in a wav and then simply convert to mp3. I am not finished yet, but Brother, I am close. Once I get it finished, I will make a vid of how I did it, and post it on my YT channel as well as my ACX Forum. Just so you know, I learned how to do this from reading your posts and replys. Thank You.

[theRamenNoodle]

Hey, guys. This is Daniel J. Lewis from The Audacity to Podcast.

-23 LUFS is not an international standard. There are multiple standards for broadcasting depending on the country: -23 LUFS, -24 LUFS, and others.

But the reason for the recommended standard of -16 LUFS (stereo) for Internet audio is that such audio is usually consumed in louder environments than broadcast media. Also, broadcast media usually needs a lot wider of a dynamic range for music and sound effects. Whereas most podcasts are spoken-word audio where there should not be as broad of a dynamic range. (If there's too much dynamic range, it's very hard to understand the words people are speaking.)

From my understanding, the -16 LUFS (stereo) originally comes from an approximation of what iTunes and other audio apps have been doing for years when you enable ondemand normalizing. Sure, it's similar to the fact that some major technologies are limited because of a horse's butt. (Short version: some tech has to fit on trains, trains go through tunnels and on tracks of certain width, tracks were laid based on previously cut trails by horses, two horses cut a particularly sized trail—it could all be a myth, but it kind of makes sense.)

Podcasts may never be regulated like radio is, but there are best practices and emerging standards that improve the audience's experience.

I agree that reaching an integrated loudness target, alone, is not good enough. It's also about loudness range (LRA) and true peak (dBTP). When you compress spoken-word audio with decent ratios, you'll end up with a smaller LRA. That prevents listeners from having to adjust playback volume within the same content. Then, the integrated loudness level is a more accurate average of the perceived loudness, with the dBTP limiter preventing clipping.

All of the reasoning aside. I think the best course of action, if the technology would support it, would be for a customizable loudness normalization tool in Audacity. This would let us set the target to -23, -24, -16, -19, or whatever other standard you wish to adhere to. Otherwise, we have a convoluted process of exporting, measuring, adjusting, and repeating; or we have to purchase other software.

But it sounds like Nyquist may be a bottleneck for this.

[steve]

Hi Daniel. good to hear from you again.

it's similar to the fact that some major technologies are limited because of a horse's butt. (Short version: some tech has to fit on trains, trains go through tunnels and on tracks of certain width, tracks were laid based on previously cut trails by horses, two horses cut a particularly sized trail—it could all be a myth, but it kind of makes sense.)

I think the best course of action, if the technology would support it, would be for a customizable loudness normalization tool in Audacity. This would let us set the target to -23, -24, -16, -19, or whatever other standard you wish to adhere to.

I agree that would be a useful tool.
We do have a plug-in that can do that for RMS level, but not yet for LUFS.

But it sounds like Nyquist may be a bottleneck for this.

The only real bottleneck, (other than someone finding the time to write it), is how much audio may be processed in one go. Without some significant changes to how Audacity and Nyquist interact, there's likely to be a maximum selection length of (guesstimation) not much more than an hour. I know that some podcasts can be longer than that, but the effect could be applied in sections. In your opinion, would a 1 hour limit on how much audio may be selected / processed in one go be a deal breaker? An error message could be thrown if the selection is too long (though this would not help if the computer has less than 2GB of available RAM - can we assume that most serious podcasters will have 2GB of available RAM?).