Re: MPEG-1 Layer 3 Question: Pause at beginning of files after decoding (not framing) (REklund )


Subject: Re: MPEG-1 Layer 3 Question: Pause at beginning of files after decoding (not framing)
From:    REklund  <roberta@xxxxxxxx>
Date:    Mon, 23 Oct 2006 11:32:54 -0700

u're welcome. -----Original Message----- From: Hermann Stemberger [mailto:hermann@xxxxxxxx Sent: Saturday, October 21, 2006 3:38 PM To: 'REklund' Subject: AW: MPEG-1 Layer 3 Question: Pause at beginning of files after decoding (not framing) Thank you very much for your help! Hermann -----Ursprüngliche Nachricht----- Von: REklund [mailto:roberta@xxxxxxxx Gesendet: Samstag, 21. Oktober 2006 22:38 An: 'Hermann Stemberger' Betreff: RE: MPEG-1 Layer 3 Question: Pause at beginning of files after decoding (not framing) hybrid MDCT has 528 sample delay plus an overlap add windowing technique w/ N=4 windowing types(short, long 2 transition windows for dynamic switching). Topic u r seeing is algorithmic delay and there should be latency/delay numbers on every compression technique out there readily available per implementation and DSP/filtebank methods utilized. Details on the various latency outs and work arounds in LAME implementation w/ some analysis of FhG encode: http://lame.sourceforge.net/tech-FAQ.txt A reasonably easy to read paper overview on MPEG 1, Layers I,II,III http://www.nue.tu-berlin.de/Publikationen/papers/nato.pdf -----Original Message----- From: Hermann Stemberger [mailto:hermann@xxxxxxxx Sent: Saturday, October 21, 2006 6:00 AM To: AUDITORY@xxxxxxxx Subject: MPEG-1 Layer 3 Question: Pause at beginning of files after decoding (not framing) Dear list, When I encode a WAV-file into MPEG-1 Layer 3 (MP3), no matter if I use the "original" patent pending Fraunhofer algorithm or the LAME algorithm I always get a few thousand sample values digital Nulls: When I decode those MP3-files again to WAV again, the signal starts 3000-4000 samples later than in the original WAV. I know about framing in MP3 - but AFAIK this should just happen to the end of the file, not to the beginning of it. There is no signal cut, the signal is just being moved forward. Does anyone know why this happens and if this is an artifact or purpose (e.g. because the decoder needs time to start)? I'm doing the whole thing for looking at simultaneous masking on one of the patent pending Fraunhofer MP3-features(not the mp3-pro). I always have to move manually the mp3-decoded file sample-accurate to the original Linear PCM-file to get the difference-signal via phase inversion of one of them. Thank you in advance, Hermann Stemberger -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.408 / Virus Database: 268.13.9/490 - Release Date: 10/20/2006 -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.408 / Virus Database: 268.13.9/490 - Release Date: 10/20/2006 -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.408 / Virus Database: 268.13.9/490 - Release Date: 10/20/2006 -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.408 / Virus Database: 268.13.9/490 - Release Date: 10/20/2006 -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.408 / Virus Database: 268.13.9/490 - Release Date: 10/20/2006 -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.408 / Virus Database: 268.13.9/490 - Release Date: 10/20/2006


This message came from the mail archive
http://www.auditory.org/postings/2006/
maintained by:
DAn Ellis <dpwe@ee.columbia.edu>
Electrical Engineering Dept., Columbia University