The title pretty much says it all. I have several mono songs that were encoded into 2-channel .m4a tracks by the Nero encoder, and would like to strip out one of the two identical channels without reencoding. How would I go about this? Thanks.
If your goal is to make the files smaller then you will probably be disappointed. Mono files encoded as stereo typically are about the same size as if they were encoded mono.
Edit: assuming vbr (quality based) encoding.
I have several mono songs that were encoded into 2-channel .m4a tracks by the Nero encoder, and would like to strip out one of the two identical channels without reencoding.
Except in special circumstances, this isn't possible because of how M/S coding works in AAC. Just leave them as stereo.
If your goal is to make the files smaller then you will probably be disappointed. Mono files encoded as stereo typically are about the same size as if they were encoded mono.
Edit: assuming vbr (quality based) encoding.
Fascinating. Slightly off the subject, does that mean a mono CD rip can be encoded to m4a in either mono or stereo at the same bitrate, with equivalent results?
Slightly off the subject, does that mean a mono CD rip can be encoded to m4a in either mono or stereo at the same bitrate, with equivalent results?
Assuming the encoder is intelligent, you'll get fairly similar results.
Fascinating. Slightly off the subject, does that mean a mono CD rip can be encoded to m4a in either mono or stereo at the same bitrate, with equivalent results?
M/S encoding works by taking the L & R channels and creating a L+R channel (mid) and a L-R (side) channel. (This process is 100% reversable, without loss.)
With a regular stereo recording, there is a lot of "center channel" sound that's common to both channels. So after M/S encoding, there is more in formation in the L+R channel than the L-R channel, and the stuff that's the same in both channels doesn't have to be encoded twice.
If you have a dual-mono recording (with the same information in both channels),
the L-R channel contains silence and the lossy encoder knows not to waste space encoding silence.If you have a stereo file with nothing in common between the channels, say English dialog in the left, and French dialog in the right, then there is no advantage, but very little harm either. (In this case, LAME would not use M/S when encoding MP3s in
Joint Stereo, and I'm assume the AAC encoder is just as smart.)
Note that with
uncompressed audio, there is no space-saving with M/S because it takes just as much space to hold 16-bits of silence as it does to hold 16-bits of actual audio. In fact, when you add/subtract two 16-bit numbers, it can take 17-bits to hold the result (depending on the value and +/- sign of the numbers being added/subtracted).
Somewhere in the audio file there will be results that don't fit into 16-bits, so it actually takes up
more space...