[Andrej Kluge]

Hi,

Is there a software that can exchange the audio channels (left/right) of
existing MP3 files, pereferably losslessly?

Reason: I have just ripped a large CD box to MP3, only to discover that the
channels are reversed. If I had known this before, I would have ripped them
to WAV, exchanged the channels and then converted them to MP3, but now this
would mean a lot of wasted time.

Of course I could just reverse the ear plugs of my headphones, but this is
no real solution...

Thanks and Ciao
AK

[Geoff]

Andrej Kluge wrote:
Hi,

Is there a software that can exchange the audio channels (left/right)
of existing MP3 files, pereferably losslessly?

Reason: I have just ripped a large CD box to MP3, only to discover
that the channels are reversed. If I had known this before, I would
have ripped them to WAV, exchanged the channels and then converted
them to MP3, but now this would mean a lot of wasted time.

Of course I could just reverse the ear plugs of my headphones, but
this is no real solution...

This is presumable some file where sides really matter ? If not, just
imagine you are in the orchestra, band, or whatever, instead of in the
audience.

geoff

[GregS[_3_]]

In article , "geoff" wrote:
Andrej Kluge wrote:
Hi,

Is there a software that can exchange the audio channels (left/right)
of existing MP3 files, pereferably losslessly?

Reason: I have just ripped a large CD box to MP3, only to discover
that the channels are reversed. If I had known this before, I would
have ripped them to WAV, exchanged the channels and then converted
them to MP3, but now this would mean a lot of wasted time.

Of course I could just reverse the ear plugs of my headphones, but
this is no real solution...

This is presumable some file where sides really matter ? If not, just
imagine you are in the orchestra, band, or whatever, instead of in the
audience.

geoff


That reminds me of a club where there was a small bar in back of the band, up on a second balcony.
It was great to listen and practically be in the band, drummer 3 foot away. Definately
the best sound in the house, rather poor out front most of the time.

greg

[Andrej Kluge]

Hi,

geoff wrote:
Of course I could just reverse the ear plugs of my headphones, but
this is no real solution...

This is presumable some file where sides really matter ?

Actually, yes. This is classical music, and if the violins play on the wrong
side it sounds definitely weird. Are there only pop/rock aficionados present
who don't understand this?

Ciao
AK

[Geoff]

Andrej Kluge wrote:
Hi,

geoff wrote:
Of course I could just reverse the ear plugs of my headphones, but
this is no real solution...

This is presumable some file where sides really matter ?

Actually, yes. This is classical music, and if the violins play on
the wrong side it sounds definitely weird. Are there only pop/rock
aficionados present who don't understand this?

Didn't you read my next sentence ;-)

But orchestral music on MP3 - surely you are joking ?

geoff

[Richard Crowley]

"Andrej Kluge" wrote ...
Is there a software that can exchange the audio channels (left/right) of
existing MP3 files, pereferably losslessly?

Reason: I have just ripped a large CD box to MP3, only to discover that
the
channels are reversed. If I had known this before, I would have ripped
them
to WAV, exchanged the channels and then converted them to MP3, but now
this
would mean a lot of wasted time.

It seems highly doubtful that it can be done *losslessly*.
Because the data for the two channels is not independent
while encoded in stereo MP3 (as it would be in WAV,
for example.)

[David Nebenzahl]

On 1/21/2010 6:33 PM Richard Crowley spake thus:

"Andrej Kluge" wrote ...

Is there a software that can exchange the audio channels
(left/right) of existing MP3 files, pereferably losslessly?

Reason: I have just ripped a large CD box to MP3, only to discover
that the channels are reversed. If I had known this before, I would
have ripped them to WAV, exchanged the channels and then converted
them to MP3, but now this would mean a lot of wasted time.

It seems highly doubtful that it can be done *losslessly*.
Because the data for the two channels is not independent
while encoded in stereo MP3 (as it would be in WAV,
for example.)

So you're saying that MP3s are encoded using deltas *between* channels,
as well as within each channel? Never thought of it that way, but it
makes sense.

Still, would it really be any more lossy than doing any kind of editing
and re-saving on a MP3 file?


--
You were wrong, and I'm man enough to admit it.

- a Usenet "apology"

[Andrej Kluge]

Hi,

Richard Crowley wrote:

It seems highly doubtful that it can be done *losslessly*.
Because the data for the two channels is not independent
while encoded in stereo MP3 (as it would be in WAV,
for example.)

Well I was hoping there would be only a bit or two to be changed in the MP3
header that control the channel assignement. (like those MP3 gain changing
programs where this is done without re-encoding the actual data)

If not, I will have to live with it.

Thanks for your reply,

Ciao
AK

[David Nebenzahl]

On 1/21/2010 11:19 PM Andrej Kluge spake thus:

Richard Crowley wrote:

It seems highly doubtful that it can be done *losslessly*.
Because the data for the two channels is not independent
while encoded in stereo MP3 (as it would be in WAV,
for example.)

Well I was hoping there would be only a bit or two to be changed in the MP3
header that control the channel assignement. (like those MP3 gain changing
programs where this is done without re-encoding the actual data)

If not, I will have to live with it.

Or you could get a 4PDT switch and connect it between amp and speakers ...


--
You were wrong, and I'm man enough to admit it.

- a Usenet "apology"

[Arny Krueger]

"David Nebenzahl" wrote in message
.com
On 1/21/2010 6:33 PM Richard Crowley spake thus:

"Andrej Kluge" wrote ...

Is there a software that can exchange the audio channels
(left/right) of existing MP3 files, pereferably
losslessly?

Not that I know of.

Reason: I have just ripped a large CD box to MP3, only
to discover that the channels are reversed. If I had
known this before, I would have ripped them to WAV,
exchanged the channels and then converted them to MP3,
but now this would mean a lot of wasted time.

Not to mention potential quality loss. I don't know if channel rotation is
described in the file header or not.

It seems highly doubtful that it can be done
*losslessly*. Because the data for the two channels is
not independent while encoded in stereo MP3 (as it would
be in WAV, for example.)

Yes and no.

So you're saying that MP3s are encoded using deltas
*between* channels, as well as within each channel? Never
thought of it that way, but it makes sense.

Both techniques are used. The sum/difference MP3s are called "Joint Stereo".

[Albie]

On Fri, 22 Jan 2010 07:44:36 -0500, "Arny Krueger"
wrote:

"David Nebenzahl" wrote in message
s.com
On 1/21/2010 6:33 PM Richard Crowley spake thus:

"Andrej Kluge" wrote ...

Is there a software that can exchange the audio channels
(left/right) of existing MP3 files, pereferably
losslessly?

Not that I know of.

Reason: I have just ripped a large CD box to MP3, only
to discover that the channels are reversed. If I had
known this before, I would have ripped them to WAV,
exchanged the channels and then converted them to MP3,
but now this would mean a lot of wasted time.

Not to mention potential quality loss. I don't know if channel rotation is
described in the file header or not.

It seems highly doubtful that it can be done
*losslessly*. Because the data for the two channels is
not independent while encoded in stereo MP3 (as it would
be in WAV, for example.)

Yes and no.

So you're saying that MP3s are encoded using deltas
*between* channels, as well as within each channel? Never
thought of it that way, but it makes sense.

Both techniques are used. The sum/difference MP3s are called "Joint Stereo".



Why does one never seem to get a straight answer around here? 8-)

AFAICS, all you need do is to play your MP3 files back via software
that offers "Reverse" channels. You're on your own there, because I
still use WinAmp 2.8 (?), and this primitive does not offer that
feature. Most dedicated pre-amps _do_ have a channel reverse.

Failing this, you require a file editor that allows Copy/Paste, but
now you must reverse each of your files. I realize you want to avoid
doing this. (Sound Forge is good, but WaveOSaur is legitimately free!
8-) Its downside is, of course, a total lack of a manual.)

As to the wag who derides playing classical music with MP3 files, he
should avoid paper, pen and ink as they say. It is well-established
by now that the vast majority of mortals can discern no difference
when the sources have been well-recorded. When differences were
statistically proved, the few golden-eared listeners (like the VP of
A&R at DG) preferred the MP3 sound to the CD!! 8-) I understand that
many heavyweights in the field of psychology were involved when MP3
was developed and tested. It is nothing to deprecate.

Oh, and I can't remember who it was, but some world-class conductor
did place his violin section on the right. 8-) Try Googling for it.
I learned this from an ancient "Classic CD" magazine, (no longer in
print... like the rag reporting the famous MP3 lab test, supra).

Albie

[David Nebenzahl]

On 1/22/2010 6:47 AM Albie spake thus:

As to the wag who derides playing classical music with MP3 files, he
should avoid paper, pen and ink as they say. It is well-established
by now that the vast majority of mortals can discern no difference
when the sources have been well-recorded. When differences were
statistically proved, the few golden-eared listeners (like the VP of
A&R at DG) preferred the MP3 sound to the CD!! 8-) I understand that
many heavyweights in the field of psychology were involved when MP3
was developed and tested. It is nothing to deprecate.

Oh, and I can't remember who it was, but some world-class conductor
did place his violin section on the right. 8-) Try Googling for it.
I learned this from an ancient "Classic CD" magazine, (no longer in
print... like the rag reporting the famous MP3 lab test, supra).

Dunno about violins on the right (don't doubt it, just have never heard
of that), but I do know that George Szell (Cleveland Orch.) swapped the
violas and celli, putting the violas on the outside right. (There's a
picture showing this on the cover of an ablum I have of his.) No doubt
there have been other unconventional seating arrangements used over the
years.


--
You were wrong, and I'm man enough to admit it.

- a Usenet "apology"

[cjt]

David Nebenzahl wrote:
On 1/22/2010 6:47 AM Albie spake thus:

As to the wag who derides playing classical music with MP3 files, he
should avoid paper, pen and ink as they say. It is well-established
by now that the vast majority of mortals can discern no difference
when the sources have been well-recorded. When differences were
statistically proved, the few golden-eared listeners (like the VP of
A&R at DG) preferred the MP3 sound to the CD!! 8-) I understand that
many heavyweights in the field of psychology were involved when MP3
was developed and tested. It is nothing to deprecate.

Oh, and I can't remember who it was, but some world-class conductor
did place his violin section on the right. 8-) Try Googling for it.
I learned this from an ancient "Classic CD" magazine, (no longer in
print... like the rag reporting the famous MP3 lab test, supra).

Dunno about violins on the right (don't doubt it, just have never heard
of that), but I do know that George Szell (Cleveland Orch.) swapped the
violas and celli, putting the violas on the outside right. (There's a
picture showing this on the cover of an ablum I have of his.) No doubt
there have been other unconventional seating arrangements used over the
years.


Nothing unconventional about that -- it's the difference between
"American" and "European" seating.

[Albie]

On Thu, 21 Jan 2010 19:06:07 +0100, "Andrej Kluge"
wrote:

Hi,

Is there a software that can exchange the audio channels (left/right) of
existing MP3 files, pereferably losslessly?

Reason: I have just ripped a large CD box to MP3, only to discover that the
channels are reversed. If I had known this before, I would have ripped them
to WAV, exchanged the channels and then converted them to MP3, but now this
would mean a lot of wasted time.

Of course I could just reverse the ear plugs of my headphones, but this is
no real solution...

Thanks and Ciao
AK


And I just have downloaded and used a program called "Audacity 1.2.6"
in the event you want to go the hard route. This editor would permit
of channel-swapping, inter alia, and the best part: It's Free!

Many tutorials available. Downloaded separately.

More of a heavyweight than WaveOSaur, but not quite as easy to use.

Albie

[Andrej Kluge]

Hi,

Albie wrote:
And I just have downloaded and used a program called "Audacity
1.2.6" in the event you want to go the hard route. This editor
would permit of channel-swapping, inter alia

On MP3 files? Without re-encoding?

(for WAV files, I already have one program which does that)

Ciao
AK

[Richard Crowley]

"Andrej Kluge" wrote ...
Albie wrote:
And I just have downloaded and used a program called "Audacity
1.2.6" in the event you want to go the hard route. This editor
would permit of channel-swapping, inter alia

On MP3 files? Without re-encoding?

No. Seems unlikely that there are ANY applications that can manipulate
MP3 without decoding and re-encoding. Same problem in the video
world with most flavors of MPEG.

[Albie]

On Sat, 23 Jan 2010 21:57:10 -0800, "Richard Crowley"
wrote:

"Andrej Kluge" wrote ...
Albie wrote:
And I just have downloaded and used a program called "Audacity
1.2.6" in the event you want to go the hard route. This editor
would permit of channel-swapping, inter alia

On MP3 files? Without re-encoding?

No. Seems unlikely that there are ANY applications that can manipulate
MP3 without decoding and re-encoding. Same problem in the video
world with most flavors of MPEG.

Au contraire! I just did it. With Audacity, one loads an MP3 file,
edits same and then "Exports" it back to Either an MP3 file or a WAV
file! It probably does silently convert to RAW in order to edit, but
the user has no inkling of it.

You might prefer the truly intuitive use of WaveOSaur. It has a
primary flaw in that it only handle file sizes of around 32 megs.
To circumvent this, download another freebie called Slice Audio File
Splitter. It will permit inaudible splits such that WaveOSaur can
accomodate them. I have used this with great success.

THIS JUST IN! I see where WaveOSaur has a "Swap Channels" function. I
have not used it, but it appears as a MenuItem under the "Process"
Menu. HEY, FINALLY A DIRECT ANSWER TO A QUESTION !! 8-)

Albie

[David Nebenzahl]

On 1/24/2010 7:03 AM Albie spake thus:

On Sat, 23 Jan 2010 21:57:10 -0800, "Richard Crowley"
wrote:

"Andrej Kluge" wrote ...

Albie wrote:

And I just have downloaded and used a program called "Audacity
1.2.6" in the event you want to go the hard route. This editor
would permit of channel-swapping, inter alia

On MP3 files? Without re-encoding?

No. Seems unlikely that there are ANY applications that can
manipulate MP3 without decoding and re-encoding. Same problem in
the video world with most flavors of MPEG.

Au contraire! I just did it. With Audacity, one loads an MP3 file,
edits same and then "Exports" it back to Either an MP3 file or a WAV
file! It probably does silently convert to RAW in order to edit, but
the user has no inkling of it.

I'm sure you did it, but this still raises the question: was your MP3
data degraded by the "silent" decoding and re-encoding that occurred
when you edited it? If so, it may not have been enough for you to have
realized that this happened.

I'm not asking this rhetorically; I actually don't know. I *suspect*
that what Mr. Crowley warned about is the case, that any application
that manipulates MP3s in any way must decode and re-encode them, which
by implication means that the data will be degraded each time, since
this is a lossy format. Is this correct?

If so, it's the same thing, by analogy, as making a xerox of a xerox of
a xerox of a xerox of a ...; each time through, the image gets a little
more degraded.

Now if only there was a bit somewhere in the MP3 header which said
"reverse channel assignments" ...


--
You were wrong, and I'm man enough to admit it.

- a Usenet "apology"

[Andrej Kluge]

Hi,

David Nebenzahl wrote:
I'm not asking this rhetorically; I actually don't know. I *suspect*
that what Mr. Crowley warned about is the case, that any application
that manipulates MP3s in any way must decode and re-encode them,
which by implication means that the data will be degraded each
time, since this is a lossy format. Is this correct?

AFAIK yes.

Now if only there was a bit somewhere in the MP3 header which said
"reverse channel assignments" ...

Exactly my notion.

I've got a promising reply from someone in the German audio newsgroup,
mentioning some internal technical specs and the need to do the programming
myself, but when there is such a possibilty I'm sure someone has implemented
it in some program somewhere.

Ciao
AK

[Richard Crowley]

"Albie" wrote ...
"Richard Crowley" wrote:
"Andrej Kluge" wrote ...
Albie wrote:
And I just have downloaded and used a program called "Audacity
1.2.6" in the event you want to go the hard route. This editor
would permit of channel-swapping, inter alia

On MP3 files? Without re-encoding?

No. Seems unlikely that there are ANY applications that can manipulate
MP3 without decoding and re-encoding. Same problem in the video
world with most flavors of MPEG.

Au contraire! I just did it. With Audacity, one loads an MP3 file,
edits same and then "Exports" it back to Either an MP3 file or a WAV
file! It probably does silently convert to RAW in order to edit, but
the user has no inkling of it.

NO, You DECODED the MP3 to WAV (or its internal equivalent) .
You may not have known that you did that because the application
did it automatically (and surreptitiously) behind the scenes. Then,
you re-encoded it back to MP3 (perhaps automatically without being
asked). But this is exactly what the OP was asking to NOT do.

[Andrej Kluge]

Hi,

Richard Crowley wrote:
Au contraire! I just did it. With Audacity, one loads an MP3
file, edits same and then "Exports" it back to Either an MP3 file
or a WAV file! It probably does silently convert to RAW in order
to edit, but the user has no inkling of it.

NO, You DECODED the MP3 to WAV (or its internal equivalent) .
You may not have known that you did that because the application
did it automatically (and surreptitiously) behind the scenes. Then,
you re-encoded it back to MP3 (perhaps automatically without being
asked). But this is exactly what the OP was asking to NOT do.

Thanks, I thought I was talking to a brick wall

Ciao
AK

[David Nebenzahl]

On 1/24/2010 1:07 PM Andrej Kluge spake thus:

David Nebenzahl wrote:

I'm not asking this rhetorically; I actually don't know. I *suspect*
that what Mr. Crowley warned about is the case, that any application
that manipulates MP3s in any way must decode and re-encode them,
which by implication means that the data will be degraded each
time, since this is a lossy format. Is this correct?

AFAIK yes.

Now if only there was a bit somewhere in the MP3 header which said
"reverse channel assignments" ...

Exactly my notion.

I've got a promising reply from someone in the German audio newsgroup,
mentioning some internal technical specs and the need to do the programming
myself, but when there is such a possibilty I'm sure someone has implemented
it in some program somewhere.

Well, it all depends on what's in the MP3 header, and from what I've
seen it doesn't look promising. Here's a page with a description of the
header:

http://www.mp3-converter.com/mp3codec/mp3_anatomy.htm

According to this, there are two fields having to do with the
channel-ness of the file:

o Channel mode (stereo, joint stereo, dual channel, single channel)
o Mode extension (used only with joint stereo, to conjoin channel data)

That second field looks intriguing, but I haven't been able to find any
sites which actually explain its function so far. Perhaps you can
determine this. But I don't think there's anything here that lets you
swap stereo channels, unfortunately.

If you could, it would be trivially easy to write a program which would
read an MP3, change some bits in the header and rewrite it. (The hardest
part would be if you had to recalculate a checksum because of the
changed data, but even this is a piece of cake.)


--
You were wrong, and I'm man enough to admit it.

- a Usenet "apology"

[Albie]

On Sun, 24 Jan 2010 12:39:21 -0800, David Nebenzahl
wrote:

On 1/24/2010 7:03 AM Albie spake thus:

On Sat, 23 Jan 2010 21:57:10 -0800, "Richard Crowley"
wrote:

"Andrej Kluge" wrote ...

Albie wrote:

And I just have downloaded and used a program called "Audacity
1.2.6" in the event you want to go the hard route. This editor
would permit of channel-swapping, inter alia

On MP3 files? Without re-encoding?

No. Seems unlikely that there are ANY applications that can
manipulate MP3 without decoding and re-encoding. Same problem in
the video world with most flavors of MPEG.

Au contraire! I just did it. With Audacity, one loads an MP3 file,
edits same and then "Exports" it back to Either an MP3 file or a WAV
file! It probably does silently convert to RAW in order to edit, but
the user has no inkling of it.

I'm sure you did it, but this still raises the question: was your MP3
data degraded by the "silent" decoding and re-encoding that occurred
when you edited it? If so, it may not have been enough for you to have
realized that this happened.

I'm not asking this rhetorically; I actually don't know. I *suspect*
that what Mr. Crowley warned about is the case, that any application
that manipulates MP3s in any way must decode and re-encode them, which
by implication means that the data will be degraded each time, since
this is a lossy format. Is this correct?

If so, it's the same thing, by analogy, as making a xerox of a xerox of
a xerox of a xerox of a ...; each time through, the image gets a little
more degraded.

Now if only there was a bit somewhere in the MP3 header which said
"reverse channel assignments" ...

As to point one, it would depend upon the listener. What is audible
degradation to one guy might pass unnoticed to another. Our channel
switcher will just have to use the free software, switch a file and
give a listen. No need to A-B test. It is personal satisfaction
only.

As to point two, someone has pointed out that one channel is but the
delta of the other. If so, a single bit switched shouldn't work. The
best answer then lies in the analog world. Like in the days of the
cave man, his preamp needs a Reverse Channels knob. 8-)

Albie

[Geoff]

Albie wrote:
As to the wag who derides playing classical music with MP3 files, he
should avoid paper, pen and ink as they say. It is well-established
by now that the vast majority of mortals can discern no difference
when the sources have been well-recorded. When differences were
statistically proved, the few golden-eared listeners (like the VP of
A&R at DG) preferred the MP3 sound to the CD!! 8-) I understand that
many heavyweights in the field of psychology were involved when MP3
was developed and tested. It is nothing to deprecate.

Me. Anything with sensitive low-level harmonicly-rich sounds in conjunction
with louder components, and many other sceanrios, is likely to suffer
audibly and obviousy to all but the most cloth-eared. But if you are happy
with the quality yu achieve, go for it.

geoff

PS, Classical music typiaclly does not have violins with flangers.

[Geoff]

Richard Crowley wrote:
"Andrej Kluge" wrote ...
Albie wrote:
And I just have downloaded and used a program called "Audacity
1.2.6" in the event you want to go the hard route. This editor
would permit of channel-swapping, inter alia

On MP3 files? Without re-encoding?

No. Seems unlikely that there are ANY applications that can manipulate
MP3 without decoding and re-encoding. Same problem in the video
world with most flavors of MPEG.

Yeah - just because it doesn't explicitly say it's not re-encoding doesn't
mean that it isn't !

geof

[Mr.T]

"Albie" wrote in message
...
As to the wag who derides playing classical music with MP3 files, he
should avoid paper, pen and ink as they say. It is well-established
by now that the vast majority of mortals can discern no difference
when the sources have been well-recorded. When differences were
statistically proved, the few golden-eared listeners (like the VP of
A&R at DG) preferred the MP3 sound to the CD!! 8-) I understand that
many heavyweights in the field of psychology were involved when MP3
was developed and tested. It is nothing to deprecate.

And anybody who claims they are unable to tell an MP3 from a wave file, and
still fails to make any mention of bit rates is just as stupid!

Fact is at 320kbs MP3's will suit most people and most music just fine. At
64kbs or less, MP3's with almost any encoder and almost any type of music is
pretty easy to pick for nearly everyone.
If *you* can't I wouldn't admit it to the world! :-)

MrT.

[Mr.T]

"David Nebenzahl" wrote in message
.com...
And I just have downloaded and used a program called "Audacity
1.2.6" in the event you want to go the hard route. This editor
would permit of channel-swapping, inter alia

On MP3 files? Without re-encoding?

If you only use it to PLAY the file and *NOT* resave it, then only one
decode to wave takes place, and no more loss than any other MP3 player
occurs.


No. Seems unlikely that there are ANY applications that can
manipulate MP3 without decoding and re-encoding. Same problem in
the video world with most flavors of MPEG.

Au contraire! I just did it. With Audacity, one loads an MP3 file,
edits same and then "Exports" it back to Either an MP3 file or a WAV
file! It probably does silently convert to RAW in order to edit, but
the user has no inkling of it.

I'm sure you did it, but this still raises the question: was your MP3
data degraded by the "silent" decoding and re-encoding that occurred
when you edited it? If so, it may not have been enough for you to have
realized that this happened.

Of *course* it adds loss when you decode AND re-encode! However what most
people don't realise is that the greatest loss occors on the first encode.
Once that part of the signal is gone, it doesn't have to be re-encoded next
time. From my experiments the first encode reduces quality by about the same
as the next 4 or 5 re-encodes. YMMV.


Now if only there was a bit somewhere in the MP3 header which said
"reverse channel assignments" ...

As others have pointed out, any player can do it easily after it has decoded
to wave. Not all MP3 players provide that facility of course, but it's
certainly not difficult to do.

MrT.

[Mr.T]

"Albie" wrote in message
...
As to point two, someone has pointed out that one channel is but the
delta of the other.

For the joint stereo mode only.

If so, a single bit switched shouldn't work.

Yes it would, all MP3 files are decoded to wave files when they are played.
The player can easily reassign which signal to output to each channel
*after* decoding takes place.
(not all players necessarily provide that facility of course, and there
appears to be no header bit to make it automatic, but it CAN certainly be
done!)


The
best answer then lies in the analog world. Like in the days of the
cave man, his preamp needs a Reverse Channels knob. 8-)

Yep, or a reverse channel software switch to do the same thing.
If he has neither he could build a line switcher or speaker switcher, or
learn to live with it and worry about something important instead.

MrT.

[Andrej Kluge]

Hi,

Mr.T wrote:
Fact is at 320kbs MP3's will suit most people and most music just
fine. At 64kbs or less, MP3's with almost any encoder and almost any
type of music is pretty easy to pick for nearly everyone.
If *you* can't I wouldn't admit it to the world! :-)

My MP3s have 256 kb/s, and this is OK for me. My hearing is not what it used
to be anyway (the trebles are dwindling), so this is -- for me -- the best
tradeoff between quality and file size.

Ciao
AK

[Doug Freyburger]

Mr.T wrote:

And anybody who claims they are unable to tell an MP3 from a wave file, and
still fails to make any mention of bit rates is just as stupid!

That or they listen to their music in their car and the background noise
overwhelms any of that detail. Why do so few folks listen to classical
music in their car? You mean there were supposed to be violins going
while that truck went by? I think the "wall of sound" concept was
introduced to rock music because so many folks were listening to the
music on their car radios so the music needed to overwhelm that truck
going by.

Side by side in an otherwise silent room, sure I can tell an MP3 from a
wave file. Ah to even have a room where I could do that on a regular
basis ...

[David Nebenzahl]

On 1/25/2010 11:09 PM Mr.T spake thus:

"David Nebenzahl" wrote in message
.com...

I'm sure you did it, but this still raises the question: was your MP3
data degraded by the "silent" decoding and re-encoding that occurred
when you edited it? If so, it may not have been enough for you to have
realized that this happened.

Of *course* it adds loss when you decode AND re-encode! However what most
people don't realise is that the greatest loss occors on the first encode.
Once that part of the signal is gone, it doesn't have to be re-encoded next
time. From my experiments the first encode reduces quality by about the same
as the next 4 or 5 re-encodes. YMMV.

I didn't know this, though it makes perfect sense now that I think of it.

It would be nice to know how this works quantitatively, based on
empirical evidence. Not doubting you, just wondering how much loss
occurs right off the bat. If the 2nd decode-encode cycle adds negligible
loss, then it might be OK to edit the file to reverse the channels.


--
You were wrong, and I'm man enough to admit it.

- a Usenet "apology"

[Dave Platt]

In article ,
David Nebenzahl wrote:

If you could, it would be trivially easy to write a program which would
read an MP3, change some bits in the header and rewrite it. (The hardest
part would be if you had to recalculate a checksum because of the
changed data, but even this is a piece of cake.)

Based on what I know/read of MPEG audio encoding, I think that the
problem you're trying to solve *is* solvable, but is rather more
complex than you would wish.

In order to swap channels in an MP3 stream, you'd need to do something
along the following lines:

- Separate the stream into frames, by detecting the frame syncs and
parsing the headers.

- Un-do the Huffman coding (a lossless operation) to recover the
frame full of perceptually-coded data (this is a lossless operation)

- Based on the information in the mode and mode extension fields,
"tear apart" the encoded (quantized) data for the left and right
channels, and then "put it back together" in the opposite
orientation. In the case of a joint-stereo encoding I imagine that
this might require swapping the sign of the inter-channel
difference value. I *think* that this swapping-around could be
done losslessly (without decoding and then re-encoding), at least
in most cases, but I can't swear to that.

- Do a new pass of Huffman encoding (which is lossless) and put the
frame headers onto each Huffman block (this might require changing
the padding bits).

- Stream out the frames.

--
Dave Platt AE6EO
Friends of Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!

[Geoff]

Doug Freyburger wrote:
Mr.T wrote:

And anybody who claims they are unable to tell an MP3 from a wave
file, and still fails to make any mention of bit rates is just as
stupid!

That or they listen to their music in their car and the background
noise overwhelms any of that detail. Why do so few folks listen to
classical music in their car? You mean there were supposed to be
violins going while that truck went by? I think the "wall of sound"
concept was introduced to rock music because so many folks were
listening to the music on their car radios so the music needed to
overwhelm that truck going by.

Side by side in an otherwise silent room, sure I can tell an MP3 from
a wave file. Ah to even have a room where I could do that on a
regular basis ...

I use 256K (non-joint) stereo MP3s in my car on long trips. I can hear the
difference easily between these and the source CDs, in the car, while
driving, and that with average pop-rock, let alone anything more subtle.

I put up with it for the convenience. In the car. Only.


geoff

[Mr.T]

"David Nebenzahl" wrote in message
s.com...
It would be nice to know how this works quantitatively, based on
empirical evidence. Not doubting you, just wondering how much loss
occurs right off the bat. If the 2nd decode-encode cycle adds negligible
loss, then it might be OK to edit the file to reverse the channels.

Simply do what I did and try it for yourself.
You won't get any "empirical evidence" here for *your* music with *your*
encoder, and *your* settings.

MrT.

[David Nebenzahl]

On 1/26/2010 8:34 PM Mr.T spake thus:

"David Nebenzahl" wrote in message
s.com...

It would be nice to know how this works quantitatively, based on
empirical evidence. Not doubting you, just wondering how much loss
occurs right off the bat. If the 2nd decode-encode cycle adds negligible
loss, then it might be OK to edit the file to reverse the channels.

Simply do what I did and try it for yourself.
You won't get any "empirical evidence" here for *your* music with *your*
encoder, and *your* settings.

No, by empirical evidence I meant some numeric value of comparison. Say,
maybe, number of samples changed between original and re-encoded stream,
or some such, as a rough measure of corruption. Of course any other
comparisons are going to be subjective. (And of course the measure of
corruption would depend on the specific source material; what I'd be
after would be an average over many samples.)


--
You were wrong, and I'm man enough to admit it.

- a Usenet "apology"

[Mr.T]

"David Nebenzahl" wrote in message
s.com...
No, by empirical evidence I meant some numeric value of comparison. Say,
maybe, number of samples changed between original and re-encoded stream,
or some such, as a rough measure of corruption.

You could have 100% of samples changed without audible corruption, (or 1
sample changed can be audible) so I can't see what you expect that to tell
you?

Of course any other comparisons are going to be subjective.

It's ALL subjective when you are dealing with lossy compression.

MrT.

[David Nebenzahl]

On 1/26/2010 9:14 PM Mr.T spake thus:

"David Nebenzahl" wrote in message
s.com...

No, by empirical evidence I meant some numeric value of comparison. Say,
maybe, number of samples changed between original and re-encoded stream,
or some such, as a rough measure of corruption.

You could have 100% of samples changed without audible corruption, (or 1
sample changed can be audible) so I can't see what you expect that to tell
you?

Well, that's true; just counting changed samples won't tell you
anything. But I'm sure there must be some way to quantify the level of
corruption, or more properly the degree of difference between a
re-encoded file and the original, based on what would actually be
audible rather than just raw numeric differences between samples. It
wouldn't be trivial to calculate, that's for sure.

Anyhow, at this point this is just an academic question and we've moved
way beyond the OP's issue.


--
You were wrong, and I'm man enough to admit it.

- a Usenet "apology"

[Peter Larsen[_3_]]

Andrej Kluge wrote:

Hi,

geoff wrote:
Of course I could just reverse the ear plugs of my headphones, but
this is no real solution...

This is presumable some file where sides really matter ?

Actually, yes. This is classical music, and if the violins play on
the wrong side it sounds definitely weird. Are there only pop/rock
aficionados present who don't understand this?

There is no reason why absolute L and R should not also matter in other
renderings of real or imaginary acoustic events.

Ciao
AK

Kind regards

Peter Larsen

[Peter Larsen[_3_]]

Richard Crowley wrote:

It seems highly doubtful that it can be done *losslessly*.
Because the data for the two channels is not independent
while encoded in stereo MP3 (as it would be in WAV,
for example.)

Depends on the mp3 options selected, however the general rule _is_ that
decode-encode would be required.

Kind regards

Peter Larsen

[Peter Larsen[_3_]]

David Nebenzahl wrote:

So you're saying that MP3s are encoded using deltas *between*
channels, as well as within each channel? Never thought of it that
way, but it makes sense.

No no no, M-S encoding is one of the options, strict X-Y encoding of high
range (discarding ramdom phase between channels) another, that one kills
real ambience.

Still, would it really be any more lossy than doing any kind of
editing and re-saving on a MP3 file?

I wouldn't worry too much on the second encode if say max quality variable
bandwidth ms-encode was the initial choice. It is also possible to select
encode as fully independent channels, but that is an unlikely initial
choice.

Kind regards

Peter Larsen