Re: lame inserting 0.047 seconds of silence in the beginning of the mp3 file
nick humphrey <[email protected]>
| Newsgroups | gmane.comp.audio.mp3.lame |
|---|---|
| Message-ID | <CAKDOLajJAdyO0_YjOgoVxwBVVaV9rH6zOwRwPAa0j=eOcEOcVQ@mail.gmail.com> |
"at least it should be obvious that it needs to bring the sample
count to a multiple of the MPEG frame size"
whatchoo talkin' 'bout willis? =)
"Are you really using mpg123 or is this mpg321 masquerading as mpg123
(this should be less common nowadays, but to make sure, please have a
look at `mpg123 --version`)?"
shell$ mpg123 --version
mpg123 1.12.1
"Input and output WAV do have the same size. That's how it should be.
Can you do that test with your setup?"
shell$ lame test.wav test.mp3
LAME 3.99.3 64bits (http://lame.sf.net)
Using polyphase lowpass filter, transition band: 16452 Hz - 17032 Hz
Encoding test.wav to test.mp3
Encoding as 48 kHz j-stereo MPEG-1 Layer III (12x) 128 kbps qval=3
Frame | CPU time/estim | REAL time/estim | play/CPU | ETA
103/103 (100%)| 0:00/ 0:00| 0:00/ 0:00| 30.900x| 0:00
--------------------------------------------------------------------------------------------------
kbps LR MS % long switch short %
128.0 1.0 99.0 73.8 15.5 10.7
Writing LAME Tag...done
ReplayGain: +1.9dB
shell$ mpg123 -w test_again.wav test.mp3
High Performance MPEG 1.0/2.0/2.5 Audio Player for Layers 1, 2 and 3
version 1.12.1; written and copyright by Michael Hipp and others
free software (LGPL/GPL) without any warranty but with best wishes
Playing MPEG stream 1 of 1: test.mp3 ...
MPEG 1.0 layer III, 128 kbit/s, 48000 Hz joint-stereo
[0:02] Decoding of test.mp3 finished.
shell$ ls -la
-rw-rw-r-- 1 nick nick 465496 2012-02-23 22:55 test_again.wav
-rw-rw-r-- 1 nick nick 39936 2012-02-23 22:53 test.mp3
-rw-r--r-- 1 nick nick 698222 2012-02-18 13:39 test.wav
test_again.wav does not have the silence in the beginning.
i compared both our outputs from lame and mpg123 and noticed nothing,
except that you are using a newer version og mpg123, which i would think
would be irrelevant.
"With plain ALSA, I can play MP3 albums ripped from CDs
that have the tracks seamlessly blended into one another and not notice
any gap."
i think ubuntu 11.10 uses ALSA under pulse as the default setup, but i cant
imagine the audio framework is of any relevance here. i might be wrong.
i'll send you the small test files i'm working with in a personal mail.
thanks for your attention so far =)
nick
2012/2/23 Thomas Orgis <[email protected]>
> Am Thu, 23 Feb 2012 17:10:39 +0100
> schrieb nick humphrey <[email protected]>:
>
> > i dont understand. the mp3 file is encoded from a wav source file using
> > *lame*. what does audacity or mpg123 have to do with it? it seems like
> > you're suggesting that the silence is normal/intended and that the
> program
> > you play it with is supposed to cut out the silence??? or have i
> > misunderstood?
>
> You understood correctly. The MP3 format includes encoder and decoder
> delays; at least it should be obvious that it needs to bring the sample
> count to a multiple of the MPEG frame size. There are technical reasons
> for this. Lame is helpful in telling how many samples it added, the
> decoder needs to interpret that (and deal with its own delay, too).
>
> Most players don't really care about that (or did in the past, at
> least), and in fact many people think MP3 files cannot be "gapless".
> But they can; I specifically worked on that feature in the mpg123
> decoder.
>
> > i run this:
> > mpg123 --loop 4 funky_drummer_1.mp3
> >
> > and the silence is there. it doesnt loop smoothly.
>
> Can you send me (personal mail, the list might not like this) the WAV
> and MP3 to verify what is going on there (I hope it's a small sample
> file)?
>
> Are you really using mpg123 or is this mpg321 masquerading as mpg123
> (this should be less common nowadays, but to make sure, please have a
> look at `mpg123 --version`)?
>
> This works for me:
>
> shell$ lame left.wav left.mp3
> LAME 3.98.2 64bits (http://www.mp3dev.org/)
> Using polyphase lowpass filter, transition band: 16452 Hz - 17032 Hz
> Encoding left.wav to left.mp3
> Encoding as 48 kHz j-stereo MPEG-1 Layer III (12x) 128 kbps qval=3
> Frame | CPU time/estim | REAL time/estim | play/CPU | ETA
> 80/80 (100%)| 0:00/ 0:00| 0:00/ 0:00| 32.000x| 0:00
>
> ------------------------------------------------------------------------------------------------------
> kbps LR MS % long switch short %
> 128.0 98.8 1.2 91.9 5.0 3.1
> Writing LAME Tag...done
> ReplayGain: -6.1dB
>
> shell$ mpg123 -w left_again.wav left.mp3
> High Performance MPEG 1.0/2.0/2.5 Audio Player for Layers 1, 2 and 3
> version 1.13.4; written and copyright by Michael Hipp and others
> free software (LGPL/GPL) without any warranty but with best wishes
>
> Playing MPEG stream 1 of 1: left.mp3 ...
> MPEG 1.0 layer III, 128 kbit/s, 48000 Hz joint-stereo
>
> [0:01] Decoding of left.mp3 finished.
>
> shell$ ls -l left*.wav
> -rw-r----- 1 thomas users 358700 23. Feb 22:18 left_again.wav
> -rw-r----- 1 thomas users 358700 11. Aug 2010 left.wav
>
> Input and output WAV do have the same size. That's how it should be.
> Can you do that test with your setup? If you have the same size but
> still a gap when looping, then perhaps it's an issue with the audio
> output code. With plain ALSA, I can play MP3 albums ripped from CDs
> that have the tracks seamlessly blended into one another and not notice
> any gap.
>
>
> Alrighty then,
>
> Thomas.
>
>
> ------------------------------------------------------------------------------
> Virtualization & Cloud Management Using Capacity Planning
> Cloud computing makes use of virtualization - but cloud computing
> also focuses on allowing computing to be delivered as a service.
> http://www.accelacomm.com/jaw/sfnl/114/51521223/
> _______________________________________________
> Lame-dev mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/lame-dev
>
>
------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/