Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Universal MG embed an audible watermark in downloads (article@post#30) (Read 62997 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Universal MG embed an audible watermark in downloads (article@post#30)

This is my first post in these forums.

Can anyone help me guess the source of some audible artefacts in some downloaded music?

I have two FLAC files for the same classical track (Tchaikovsky's 5th Symphony, Markevitch, on a Philips label). One was purchased from Passionato, the other was purchased from UMG's web site. Although both are supposed to be lossless copies of the data on the original CD, the FLAC file from Passionato actually has a clearly audible sort of "flutter" in some quiet, sustained passages. It is not the same as the UMG file. The difference isn't subtle. (Having read the TOS, I did do an ABX test. But it hardly seemed necessary in this case: I got 10/10 without difficulty. I will upload the short samples that I used.)

My guess was that the Passionato download was actually derived from a compressed audio file, converted to FLAC. But I'm puzzled. Compressing the UMG file and then converting back to FLAC did not give me a result that was in any way comparable, whatever compression I tried. The UMG and the Passionato file have exactly the same number of samples, and the difference of the two wave forms has a spectrum which is mostly in the range of about 1000 Hz -  2500 Hz, with very little above that.

Another possibility is that both FLAC files were obtained from some (fairly high quality) compressed source -- say MP3 -- but that the Passionato file was decompressed using software with a bug. I have quite a few FLAC files from Passionato. All the files from UMG labels exhibit this "flutter", with the exception of the most recent one I purchased.

I will edit this post when I have a link to the samples. I'll upload 7 seconds of each of the two FLAC files, and also the difference of the two wave forms (amplified). The difference sounds like fragments of the original tones together with higher "chirps". (I read the TOS concerning differences of wave forms, but I think the difference is interesting in this case.)

Edit: The uploaded FLAC samples are here.

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #1
Interesting town name you choosed
With classical recordings you experience the most strange versions. I only can speak for some classical RCA recordings that were released as first release, at least 2x re-release in japan followed by remastered versions and now after sony took them over another version under their label!
I don´t know it helps here but your 2 versions may come from 2 very different masters. Judging the noisefloor of both my bet is on both being lossless. You may have the chance to get more info out of the bootlegs coming with the releases. Sometimes they name the mastereing engineers or studio.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #2
After looking at the two samples it strongly appears as if they were decedents from the same digital source.

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #3
Hmm, interesting, and I agree with Wombat (edit: and greynol), I think they both come from true lossless masters, as well.

The reason I don't think the Passionato track is sourced from lossy is that, in addition to what Wombat noted about the noise floors, the fluttering is fairly slow, but very rhythmic and consistent, which is usually not the case with fluttering/ringing artifacts in lossy encoding.  Yes, those artifacts can be somewhat rhythmic, but they're usually much faster.

I was going to guess wow and flutter (specifically wow) from an analog tape machine, but if this is the release you're referring to, it appears to be all-digital, so that couldn't be the case (and all that hiss would then be coming from the mics and/or preamps).

Odd indeed...was there any liner note info included with the Passionato download WRT the mastering engineers or facilities?
"Not sure what the question is, but the answer is probably no."

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #4
Hmm, interesting, and I agree with Wombat (edit: and greynol), I think they both come from true lossless masters, as well.

The reason I don't think the Passionato track is sourced from lossy is that, in addition to what Wombat noted about the noise floors, the fluttering is fairly slow, but very rhythmic and consistent, which is usually not the case with fluttering/ringing artifacts in lossy encoding.  Yes, those artifacts can be somewhat rhythmic, but they're usually much faster.

I was going to guess wow and flutter (specifically wow) from an analog tape machine, but if this is the release you're referring to, it appears to be all-digital, so that couldn't be the case (and all that hiss would then be coming from the mics and/or preamps).

Odd indeed...was there any liner note info included with the Passionato download WRT the mastering engineers or facilities?


Thanks for these observations. The release isn't the one you linked to: it is this one. It was definitely an analog recording (from the 1960s). I agree with greynol: the two FLAC files clearly have a common digital ancestor somewhere down the line, because there's no way that the data could be so close otherwise.

If it was restricted to one instance, then maybe it might be a question of a different mastering. But the same problem is present on (nearly) all the downloads that I have from passionato that are on labels from UMG. Also, although the flutter is only clearly audible in certain passages (e.g. sustained woodwinds playing mezzo-piano), it is so obvious there that I can't believe it was ever present on an officially-released disk.

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #5
It could be the result of noise reduction or perhaps some type of watermark.  I don't see anything obvious indicating lossy compression.

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #6
It could be the result of noise reduction or perhaps some type of watermark. ...


Ahh.

The noise reduction idea seems a bit unlikely in the circumstances (applied to all FLAC files in passionato's store). But watermarking would make sense (UMG adding a watermark to all the files they provide to passionato.com).

I wonder, could the effects of watermarking really be that bad?

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #7
I was going to guess wow and flutter (specifically wow) from an analog tape machine


That was my thought too.
daefeatures.co.uk

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #8
I was going to guess wow and flutter (specifically wow) from an analog tape machine


That was my thought too.


I do have some entirely digital recordings from passionato.com with the same problem. For example, I have this version of Mahler 9, in FLAC format, and it has the same sort of "flutter" effect (e.g very apparent in the horn phrases at 00:35 to 00:45 in the first movement). For this release, I do not have the corresponding FLAC file from the Universal Music web site (the link above), so I can't make a comparison. But I have listened to the clip of this track that is available on Amazon's MP3 web site. Although the streaming quality of the clip is low, it is good enough to be fairly sure that the flutter is not present.


Universal MG embed an audible watermark in downloads (article@post#30)

Reply #9
One thing i noticed is the noise floor in the delta file looks like noise shaped dither. So one of these recordings must have gone thru some digital processing or downsampling different as the other file.
I wonder if one version was processed to correct the flutter. Imho this is pretty hard to repair. The other way around makes not much sense but who knows!?
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #10
I guess I am leaning to the explanation suggested by greynol: some sort of watermarking. Reading around, it seems that Universal Music do add watermarks to the files they provide to online music stores. And perhaps heavy-handed watermarking might lead to these artefacts.

I think the "flutter" is audible on some of the track sample clips on passionato's web site:
  • The Mahler 9 version that I mentioned above. Listen to the french horn after 35 seconds or so (about two thirds of the way through the clip).
  • This recording of the Brahms piano concerto no. 2: list to the horn, particularly the third note of the first phrase and last two notes of the second, falling phrase. Maybe a very nervous horn player, but I don't think so. The piano shows the same problem sometimes.

I don't have that Brahms recording, but I do have the Mahler in FLAC format from passionato (as mentioned in my previous post), and the FLAC version seems to be the same as far as the flutter goes.

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #11
I guess I am leaning to the explanation suggested by greynol: some sort of watermarking. Reading around, it seems that Universal Music do add watermarks to the files they provide to online music stores. And perhaps heavy-handed watermarking might lead to these artefacts


If this comes thru to be true i´d say this is scandalous!! Why to pay for a lossless version that adds much more noise as any modern lossy encoding does?
Time to write an Email and complain directly at them asking for clarification. Will be interesting how they explain it to you.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #12
It also seems that passionato's files from other labels are not affected (or not all of them anyway). The second movement of Brahm's 4th symphony begins with horns then clarinets.

These clips are low bitrate MP3 obviously, so there may be other factors at play. I have the original CD for the first of those two Brahms clips, and the horns and clarinets are -- of course -- completely smooth.

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #13
So, because this is such a bizarre artifact, and because I'm also incredulous that an online music store with such an extensive catalog could be screwing things up this badly with lossless tracks, I decided to just pick a track at random that I could also find in FLAC form on Deutsche Grammophon's and download both.

I'm not much of a fan of orchestral or symphonic music (I consider The Beatles to be "classical" music  ), but I do enjoy medieval and Renaissance music every now and then, so I just picked an album at random - Cristóbal de Morales: Music for Philip II - Requiem - and checked a few of the samples on Passionato's website, to see if there was any sign of that vibrato-like artifact even in the samples.

Sure enough, it very much sounded like there was, so I picked track 07 - where it seemed very pronounced - and downloaded it from both Passionato and UMG.

I posted the first 15 seconds of each file to mannheim's original upload thread.

As the kids say these days:  O...M...F...G.

The music is just massed male voices, and the vibrato artifact is incredible, especially around the 0:09 mark.

As expected, the UMG file is perfectly clean, and while I'll leave it to the end user to do an inverted mix paste in his or her audio editor of choice, the remaining artifacts are identical in sound to the ones created from mannheim's sample files.

If that is indeed watermarking being applied by Deutsche Grammophon, it's shameful - indeed, scandalous, as Wombat said - and whether it's their fault or not, Passionato really has no right calling those tracks lossless.
"Not sure what the question is, but the answer is probably no."

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #14
If this comes thru to be true i´d say this is scandalous!! Why to pay for a lossless version that adds much more noise as any modern lossy encoding does?
Time to write an Email and complain directly at them asking for clarification. Will be interesting how they explain it to you.


Well I don't want to rush to judgement. I did send three emails to Passionato, without any reply. If there is a watermarking problem, then I think it is more Universal's fault. But perhaps it's just me and I've inadvertently done something stupid? [Edit: Okay, I just read mixminus1's post, above; so I'm probably not crazy.] The only other thing I've found on the web that says someone else has had this problem is this thread on computeraudiophile.com. The orignal poster over there found the problem present in files from passionato but also from iTunes. All the tracks that are mentioned in that thread are on labels belonging to Universal.

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #15
Oh, you're definitely not the one who's done something stupid here.

I just checked the samples of the Morales album on iTunes, and I'm not hearing any signs of that vibrato artifact.

Regardless of who's at fault, it's absurd for "legitimate" downloads - lossless or otherwise - to have this artifact added to them.  Not to put too fine a point on it, but I think it's safe to say you'd never find this kind of f**k-up on FLAC files "in the wild."
"Not sure what the question is, but the answer is probably no."

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #16
Sure enough, it very much sounded like there was, so I picked track 07 - where it seemed very pronounced - and downloaded it from both Passionato and UMG.

I posted the first 15 seconds of each file to mannheim's original upload thread.

As the kids say these days:  O...M...F...G.

The music is just massed male voices, and the vibrato artifact is incredible, especially around the 0:09 mark.

Good find and easy to spot the "vibrato" effect. So your brought up samples, mannem did lead to something. Time to nag Passionata again. (mannem is mannheim for native speakers )
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #17
Since the vibrato artifact was just as present in the streaming sample as it was in the FLAC file (also noted by mannheim), I clicked through another dozen or so samples of choral music from other labels, and never once heard a trace of the artifact.

On a whim, picked another DG release - Palestrina sung by the Westminster Abbey Choir - and clicked on a sample:  sure enough, about a third of the way in, there was the vibrato, clear as day.

So, while this is very much looking to be specific to Deutsche Grammophon (at least, and perhaps other UMG labels, as well), Passionato isn't necessarily in the clear yet WRT the degraded audio - we'd need to know if they did any additional processing to the files they received from DG, and hopefully mannheim receives an (informative) answer from them.

Edit: added links
"Not sure what the question is, but the answer is probably no."

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #18
Many thanks to all of you here for helping me confirm this problem and narrow it down. I'll post back again when I hear from Passionato.

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #19
Many thanks to all of you here for helping me confirm this problem and narrow it down. I'll post back again when I hear from Passionato.



@mannheim: I've asked our customer service person to get back to you. Apologies for the delayed response. Our resources have been stretched very thin over the past several months as we're in the process of making some major changes to Passionato.

W/r/t the Universal FLAC issue, Universal audio files purchased  from Passionato are received as FLAC files from UMG. Not all distributors offer their audio to us in FLAC (some offer WAV or AIFF and we then convert to FLAC). I strongly suspect the audio artifacts you are hearing are present in the files we've received from UMG. I will try to pursue an answer from someone within Universal but I suspect a response will be unlikely. If I do hear back, I will post a response here.

Thank you,
Jim Willis
Passionato


Universal MG embed an audible watermark in downloads (article@post#30)

Reply #20
Jim, thank you very much for responding here.

Given that UMG has (apparently) willfully supplied you with degraded audio files, I won't be holding my breath for a response, either, but thanks again for posting.
"Not sure what the question is, but the answer is probably no."

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #21
Possibly related: there's some information here about "eFolio":

Quote
eFolio Watermark tool is the official file transferring tool for Universal Music Group. This service will transfer and watermark .mp3, .wav, .m4a and .aiff audio files


Hard to tell if this is still current information.

Then there's this, where the writer claims that eFolio watermarking is audible as a sort of "warble". Seems to fit, but who knows.

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #22
Since this was just covered by TechDirt (UMG Watermarks Audiophile Files, Pisses Off Paying Customers), I'm curious if there have been any further developments in confirming this is UMG watermarking, and getting a position from Passionato (whom I think may be at more risk from all of this than UMG).

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #23
I never heard back from passionato.

Universal MG embed an audible watermark in downloads (article@post#30)

Reply #24
I believe that the same thing may have just happened to me, looking for the best legal (finding a FLAC download link proved impossible) FLAC quality download of Amy Winehouse's "Back to Black" I eventually navigated myself to https://store.universal-music.co.uk where it came to my surprise that FLAC downloads were available. Upon listening to the FLAC files I'd purchased I noticed terrible quality, it sounded as if they did not even take the time to rip a cd copy to 16bit FLAC, it just sounded like a terrible MP3, I found many artifacts in each track, and at many points a distorted sound was evident. I don't know if this is watermarking or just terrible audio quality, but being that this site is in the family of UMG I'm sure the problem is of the same nature. I just e-mailed them, although I'm sure I'll never get a response...