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: A way to tell if a mp3 is stereo or joint stereo? (Read 25474 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

A way to tell if a mp3 is stereo or joint stereo?

Is there a program or some way to tell if a mp3 was made stereo or joint stereo? I'm not too informed on mp3s yet, sorry if this is a weak question. Anyway, how can i tell?

A way to tell if a mp3 is stereo or joint stereo?

Reply #1 –
Try Googling for a free version of EncSpot. That'll tell you.

Cheers, Slipstreem. 

A way to tell if a mp3 is stereo or joint stereo?

Reply #3 –
Mp3GuessEnc is a good command-line tool with the same functionality as EncSpot.

A way to tell if a mp3 is stereo or joint stereo?

Reply #4 –
You do know, don't you, that all else being equal, joint stereo is preferable over stereo?

A way to tell if a mp3 is stereo or joint stereo?

Reply #5 –
Quote from: pdq on
You do know, don't you, that all else being equal, joint stereo is preferable over stereo?
That all comes down to personal preference.

A way to tell if a mp3 is stereo or joint stereo?

Reply #6 –
Quote from: Netherworld on
That all comes down to personal preference.

Not really. Most types of joint-stereo losslessly store data in a way that makes files smaller.

All other things being equal, any file using simple L/R stereo over joint-stereo will sound exactly the same and be bigger.
elevatorladylevitateme

A way to tell if a mp3 is stereo or joint stereo?

Reply #7 –
Or, in the case of CBR, be the same size but have inferior quality.

A way to tell if a mp3 is stereo or joint stereo?

Reply #8 –
First of all, im a new user here at HA! I´ve read lots of topics for some time now, but never tried to register and start posting here. Finally i did it. Ill try look for a "newbies go here" topic and read further faq/rules posts, apart from the basic 13 rules prior to registering, for good posting afeter i do this little comment^_^.

Quote from: shakey_snake on
Quote from: Netherworld on

That all comes down to personal preference.

Not really. Most types of joint-stereo losslessly store data in a way that makes files smaller.

All other things being equal, any file using simple L/R stereo over joint-stereo will sound exactly the same and be bigger.



Well i have read lots of topics on hidrogean audio for a long time now, im not a hardcore audiophille who follows every detail about aundio enconding and releases, neither do i own expensive high-end audio phones or sound cards, but anw, i understand a little bit of the gerenal. Recently specially i´ve read lots of topics talking about how Joint stereo is better in all spects over regular stereo and that it is supposed to be a lossless mask with not noticeable artifacts... (I do know that is a myth that joints loses stereo information)

HOWEVER, for years now, i´ve always noticed more artifact on high pitchs drums in rock music that is encoded with joint stereo. Of course that could be mere subjectivism, and is truth that i´ve already tried ABX between a joint-stereo and regular stereo with the same bitrate (abr), and had difficult inconclusive results between the two samples, even tough i could easyly distinguish the artifacts from the original file samples that i used when encoded at 128~180kbps.

However my doubt here is that i can indeed easily noticea diff betwen a joint and a regular stereo when using the "karaokê" option in my onboard sound card driver software:



EDIT:
I´ve just tried to make an mid-center discard using sound forge right now, with two 128cbr samples that i used there, one joint and another regular stereo for test, and the edited mid-stripped mono from tru stereo sample sounded with lots less artifacts, so as i tought, is not my hardwares fault. I can post the samples here if anyone is interested. If im doing something stupid or absurd please tell any1 tell me. o.o

A way to tell if a mp3 is stereo or joint stereo?

Reply #9 –
Quote from: sephirotic on
Any toughts? i know is a little bit offtopic, but i think is still a "valid" post if not, please tell me and ill check a more proper one to continue this discussion....
It seems you are having a hard time "letting go" of some sort of bias you have against joint stereo.

example:
Quote
and is truth that i´ve already tried ABX between a joint-stereo and regular stereo with the same bitrate (abr), and had difficult inconclusive results between the two samples, even tough i could easyly distinguish the artifacts from the original file samples that i used when encoded at 128~180kbps.
...makes absolutely no sense at all.

If your ABX results were inconclusive, then that means that you COULD NOT "easyly distinguish the artifacts from the original file samples". Period.

As for "karaokê mode" goes, we have no idea what that does to the waveform at playback (I assume it alters the output waveform in an attempt to remove the lyrics). If you can indeed conclusively ABX a difference with "karaokê mode" enabled, then the only thing you should assume is that your soundcard software is poorly written, which isn't particularly uncommon for bundled soundcard software.
elevatorladylevitateme

A way to tell if a mp3 is stereo or joint stereo?

Reply #10 –
Quote from: shakey_snake on
If your ABX results were inconclusive, then that means that you COULD NOT "easyly distinguish the artifacts from the original file samples". Period.

As for "karaokê mode" goes, we have no idea what that does to the waveform at playback (I assume it alters the output waveform in an attempt to remove the lyrics). If you can indeed conclusively ABX a difference with "karaokê mode" enabled, then the only thing you should assume is that your soundcard software is poorly written, which isn't particularly uncommon for bundled soundcard software.




As i said, i do kwow that hardware processing, specially this one for karaoke, is not a reliable way of  comparative "testing".

however as i mentioned on my edit in my topic, i then used a professional audio editing software for excluding the "center" information of two 128cbr samples. One stereo and another in joint, and the result was the same, clearly perceptive, not even abx was needed so is not my hardware fault....

Here are the samples...

http://rapidshare.de/files/39735635/stereo_vs_joint.7z.html
(is rapidshare allowed? i must read the rules ASAP, anyway if not, ill edit later)

Yes i know that in the end what matter is the "individual subjective perceptive quality" however listen those samples, they are CLEARLY different from each other. Even tough they are edited from the original sound they mean that there is the artifacts even tough "masked" in the the original files, so MAYBE it is possible to hear the difference between both files via abxing. I KNOW IM TALKING ABOUT POSSIBILITIES, however, if the joint-stereo really was a LOSSLESS method of processing stereo information, then why should be more artifcat in the waveform of the joint-stereo file? I used sound forge CHANNEL processing tool, is there the problem, or in the decode of the mp3, or it isnt lossless at all? is just lossy? THAT is my point here and what im trying to figure out.

Ill do another ABX later and post my results about it here later. However, what i expect is, as always, perceive easily artifacts, but have a hard time telling wich one is more annoying, that is harder to do. However, with the FILTERED samples, is obvious that the joint is more annoying.

A way to tell if a mp3 is stereo or joint stereo?

Reply #11 –
What you are discovering is that lossy codecs are designed to sound as close as possible to the original when listening as normal stereo at normal volume levels. The objective is to hide artifacts the best way possible given this listening environment.

There are atypical ways of listening to these files, such as surround-sound and, apparently, karaoke, for which artifacts that are normally inaudible become audible. Under these conditions all lossy codecs are likely to fail to some degree, and it is possible that joint stereo could actually worsen the problem.

A way to tell if a mp3 is stereo or joint stereo?

Reply #12 –
Quote from: sephirotic on
however as i mentioned on my edit in my topic, i then used a professional audio editing software for excluding the "center" information of two 128cbr samples. One stereo and another in joint, and the result was the same, clearly perceptive, not even abx was needed so is not my hardware fault....


WTF.
You remove data from the file and you wonder what it sounds worse?!?
Of course it would sound worse after YOU'VE REMOVED NECESSARY AUDIO DATA and effectively crippled the file.

But why under any circumstances would you need to remove this "center" (ie Mid-side Stereo) information? 
Short answer: you wouldn't.
So you've found no differences.
elevatorladylevitateme

A way to tell if a mp3 is stereo or joint stereo?

Reply #13 –
Quote from: shakey_snake on
Quote from: sephirotic on

however as i mentioned on my edit in my topic, i then used a professional audio editing software for excluding the "center" information of two 128cbr samples. One stereo and another in joint, and the result was the same, clearly perceptive, not even abx was needed so is not my hardware fault....


WTF.
You remove data from the file and you wonder what it sounds worse?!?
Of course it would sound worse after YOU'VE REMOVED NECESSARY AUDIO DATA and effectively crippled the file.

But why under any circumstances would you need to remove this "center" (ie Mid-side Stereo) information? 
Short answer: you wouldn't.
So you've found no differences.


Wrong, i did. Just did a ABX test between both samples.

And i didn´t removed "NECESSARY AUDIO DATA", i removed CENTER AUDIO SIMILARITY, BETWEEN TWO SIMILAR AUDIO-FILES. Due to the nature of the joint-stereo architeture your argument is falacious, not mine. THERE SHOULDN´T BE artifact in the joint stereo if the process is lossless (as far as i know). U didn´t argumented AGAINST that.

here is my abx of the original files:

Quote
foo_abx 1.3.1 report
foobar2000 v0.9.5.3
2005/06/16 10:15:17

File A: I:\Musicas\Classic standard prog rock\Offspring\testes\stereo vs joint\joint-ste128.mp3
File B: I:\Musicas\Classic standard prog rock\Offspring\testes\stereo vs joint\true-ster128.mp3

10:15:17 : Test started.
10:16:50 : 01/01  50.0%
10:17:08 : 02/02  25.0%
10:17:32 : 03/03  12.5%
10:17:58 : 04/04  6.3%
10:18:21 : 05/05  3.1%
10:18:41 : 06/06  1.6%
10:18:56 : 07/07  0.8%
10:19:11 : 08/08  0.4%
10:19:18 : 09/09  0.2%
10:19:46 : 10/10  0.1%
10:19:52 : 11/11  0.0%
10:20:00 : 12/12  0.0%
10:20:08 : 13/13  0.0%
10:20:16 : 14/14  0.0%
10:20:26 : 15/15  0.0%
10:20:53 : 16/16  0.0%
10:20:58 : 17/17  0.0%
10:21:07 : 18/18  0.0%
10:21:19 : 19/19  0.0%
10:21:51 : 20/20  0.0%
10:21:59 : Test finished.

----------
Total: 20/20 (0.0%)


Yes, i managed to quickly, and easily distinguish via ABX, 2  encoded lame mp3 at q0, 128kbps CBR, one joint-stereo and another regular stereo.

The raw-wav sample was a of pop-punk rock (specifically, a excert of Staring on the sun from offspring)

Pal, maybe im not expressing myself very well. as i´ve read in other posts, joint stereo are supposed to be a lossless way to encode stereo data. This way, no quality loss is archived in this method IN ANY WAY, thus saving bitrate for mid informating and so on, improving quality of a file in a certain bitrate if it was  encoded with regular stereo separation.

However, i always tough that joint stereo wasnt lossy and i tought that affirmation was ODD. Of course im not a stupid scammer to say: "HEY! THE LAME DEVELOPER IS WRONG! JOINT STEREO SUX!".

lol

So here im with an ABX test and samples from my computer of a joint and a stereo mp3 files.

And im here humbling asking what i did wrong if joint stereo is supposed to be lossless AND BETTER than regular stereo. Im not trying to say everyone is wrong, of course i believe in the experts, but for some reason, for me, joint stereo never was the same thing. And here are the ABX and the samples to prove it wasn´t just my subjectivism. So if it is not, i want to know what is it.

And u haven´t told me so far that, only trying to prove that im wrong and everything is my subjectivism

it´s not.

Of course probably im wrong with the joint-stereo stuff, but is not the samples here my problem. that what i want to know.

Did u get it now?

=P

here are the samples i used in the abx, plus the original raw if you suspect my encoded mp3 files were at flaw:

http://rapidshare.com/files/122845654/ster...joint2.zip.html

A way to tell if a mp3 is stereo or joint stereo?

Reply #14 –
Joint stereo encoding IS lossless (other than rounding errors) - it is the lossy encoding which follows that is not lossless. The lossy encoding can result in differences when encoding mid/side vs. left/right, and for normal listening when the encoder chooses mid/side it is because the result is either smaller size or more accurate reproduction. For atypical listening conditions all bets are off because that is not what the codec was tuned for.

A way to tell if a mp3 is stereo or joint stereo?

Reply #15 –
Quote from: pdq on
Joint stereo encoding IS lossless (other than rounding errors) - it is the lossy encoding which follows that is not lossless. The lossy encoding can result in differences when encoding mid/side vs. left/right, and for normal listening when the encoder chooses mid/side it is because the result is either smaller size or more accurate reproduction. For atypical listening conditions all bets are off because that is not what the codec was tuned for.



Quote from: pdq on
What you are discovering is that lossy codecs are designed to sound as close as possible to the original when listening as normal stereo at normal volume levels. The objective is to hide artifacts the best way possible given this listening environment.

There are atypical ways of listening to these files, such as surround-sound and, apparently, karaoke, for which artifacts that are normally inaudible become audible. Under these conditions all lossy codecs are likely to fail to some degree, and it is possible that joint stereo could actually worsen the problem.


Hello there pdg, thanx for the polite answer o/

yes i realize, that. However, as i´ve mentioned, recently i was reading about joint stereo and regular stereo methods of encoding.

I ALWAYS tought, that JOINT-STEREO was a LOSSY METHOD for saving bitrate thus improving quality. But in  my allienated oppinion i always tought joint-stereo method had to many artifacts and endend not being a good method, so i always encoded with stereo-only

However, recently as i was reading about the subject, i read this topic:

http://www.hydrogenaudio.org/forums/index....c=30637&hl=

In this topic a LAME DEVELOPER says, that the Joint-stereo method SHOULDN`T generate artifacts because is a LOSSLESS algorythm of stereo handling is others factors that generate that, but the gain in quality should be compensatory.



So i got confused and started doing these tests mentioned here with samples to see if it was my subjectivism.

However in my previous sample, you can hear MORE artifacts, no less. it wasn´t subjectivism.

i wanted to know why i found those artifacts, if i used regular cbr encode options and they weren´t supposed to be there like i previously read.

A way to tell if a mp3 is stereo or joint stereo?

Reply #16 –
Quote from: sephirotic on
And i didn´t removed "NECESSARY AUDIO DATA", i removed CENTER AUDIO SIMILARITY, BETWEEN TWO SIMILAR AUDIO-FILES.

All data is necessary for a psycho-acoustic (lossy) file to function properly, even center-channel similarity.
As soon as you remove data from lossy files, you are ABXing the crippled files you've created, but that has nothing to say about the uncrippled version of the files.

What your most recent ABX'ing may have shown, is that joint-stereo lossy files do not transcode as well as L/R "regular stereo". But, who trascodes from a lossy source anyways?

Quote
Due to the nature of the joint-stereo architeture your argument is falacious, not mine. THERE SHOULDN´T BE artifact in the joint stereo if the process is lossless (as far as i know).
If you've bothered to read the wiki page I've linked to above, the mid-side stereo-type of joint-stereo is lossless. Intensity-stereo is a psycho-acoustic (ie lossy) form of joint-stereo, but is typically only used at very low bitrates.

Notice specifically what I've said in my first post in this thread:
Quote from: shakey_snake on
Not really. Most types  of joint-stereo losslessly store data in a way that makes files smaller.

All other things being equal, any file using simple L/R stereo over joint-stereo will sound exactly the same and be bigger.

Now, the fact that you still cannot ABX non-crippled files proves my underlined point.
elevatorladylevitateme

A way to tell if a mp3 is stereo or joint stereo?

Reply #17 –
Quote from: shakey_snake on
Now, the fact that you still cannot ABX non-crippled files proves my underlined point.




No, the ABX i did there was for the INTACTED mp3 files, not the crippled ones from before. Your point is still at flaw.

I COULD ABX REGULAR mp3 SAMPLES AND DISTINGUISH BETWEEN JOINT AND-NON JOINT STEREO FILES. REREAD the post #14.

however i understand the rest of your arguments now.

(PS: OMG im so confused with the EDIT and QUOTE system of this forum o.o, im trying to fix my posts now, please excuse me for the mess)

A way to tell if a mp3 is stereo or joint stereo?

Reply #19 –
Quote from: shakey_snake on
Quote from: sephirotic on

i removed CENTER AUDIO SIMILARITY, BETWEEN TWO SIMILAR AUDIO-FILES.

...therefore you crippled them.


as i said, REREAD THE POST #14,

i used the original sample for the ABX not THE CRIPPLED ONES!!!!

geez!

i didnt use the karaoke effect neither lol.

If u still don´t believe/understand what im saying (sorry for my bad english) download this:

Quote
here are the samples i used in the abx, plus the original raw if you suspect my encoded mp3 files were at flaw:

http://rapidshare.com/files/122845654/ster...joint2.zip.html

A way to tell if a mp3 is stereo or joint stereo?

Reply #20 –
No one said that stereo and joint stereo files will sound the same if they are CBR. If artifacts are audible in one then there will probably be artifacts in the other. What is being said is that all else being equal, the joint stereo file should sound closer to the original than the stereo file. To ABX the two lossy files against each other proves nothing. What you want here is ABC/HR testing.

A way to tell if a mp3 is stereo or joint stereo?

Reply #21 –
Quote from: pdq on
No one said that stereo and joint stereo files will sound the same if they are CBR. If artifacts are audible in one then there will probably be artifacts in the other. What is being said is that all else being equal, the joint stereo file should sound closer to the original than the stereo file. To ABX the two lossy files against each other proves nothing. What you want here is ABC/HR testing.


Yes i do know that. there will be notable artifacts in most mp3 at 128kbps indepedent of they being joint or regular stereo.

However in my little understanding about this (and that was said also in the topic i linked here), is that in the same bitrate, a joint stereo file, SHOULD sound better (with a little bit LESS artifacts) than a regular stereo at the same size/bitrate due to the nature of bitrate economy of the JS.

Both files i abxed had artifacts. And still i could perfectly distiguished between then. That is because one of then had Notable MORE artifacts than the another.

More artifacts (generally) means "less close" than the 'original file', so, lesser quality. (right?)

And the one that had more artifacts in my samples was the joint stereo, not the opposite.

So in MY SAMPLES, this mean the joint was inferior..
that´s the whole point, whats wrong with those samples? If someone believe that my test had any flaws with the samples, then that person must do the abx again with new samples. If the result ends up been similar, then the whole point of my series of posts:

What is the explanation for that?

-----------------------
Is that what you were trying to say?

Anyway what is an ABC/HR test?

A way to tell if a mp3 is stereo or joint stereo?

Reply #22 –
Quote from: sephirotic on
That is because one of then had Notable MORE artifacts than the another.

ABX testing does not prove this, it only shows differences between the two.

Therefore your statement:
Quote from: sephirotic on
And the one that had more artifacts in my samples was the joint stereo, not the opposite.

...is completely unproven.
elevatorladylevitateme

A way to tell if a mp3 is stereo or joint stereo?

Reply #23 –
Quote from: shakey_snake on
Quote from: sephirotic on

That is because one of then had Notable MORE artifacts than the another.

ABX testing does not prove this, it only shows differences between the two.

Therefore your statement:
Quote from: sephirotic on
And the one that had more artifacts in my samples was the joint stereo, not the opposite.

...is completely unproven.


EDIT:

Errr... i disaggre. The abx doesnt prove with one is better, actually the abx doesnt do NOTHING. It only prove im not in my subjectiviness giving false statements. The ABX did prove that my perception of one been different from another is true! "Difference" means they have 'different aspects'. If u can notice a difference, you can notice WHAT that difference is. What i notice is: One sample has slightly more high-pitch artifacts than the another. I can´t prove that too you, actually i can´t prove nothing. You can chose on always "relativate" one affirmation of my to brake my thesis or choose to accept then and debate what could have gone wrong.

Even so, if you don´t agree with the previous logical paragraph statemen The previous CRIMPLLED samples i put here does  prove NOTABLY the difference in artifacts between the two samples


Of course the ABX only proves i can objectively DISTINGUISH both files.
I do know that the fact i can distinguish is one thing, but the fact i like one better than another still could be subjective. THAT i do agree.

However, thats why i did the crimpling before, to clearly show wich one had more artifacts than the another.

A way to tell if a mp3 is stereo or joint stereo?

Reply #24 –
Quote from: sephirotic on
Errr... yeah, the ABX doesnt prove that, but the previous CRIMPLLED samples i put here does 

How exactly?

You need to realize that no matter how many times you've asked me to reread post 14, I still, apparently, have no idea what your exact methodology was.

[edit]dammit quit editing your posts.
This is the most confusing forum conversation I've ever been apart of.
elevatorladylevitateme