Hi. Since I'm new to this mailing list, I don't know whether this was discussed in this mailing list, or if it is already fixed. Anyway, the problem is that all audio encoders out there add a small amount of silence to the start/end of the encoded file. This is usually a small amount, around 0.03 seconds. Most of the time it goes unnoticed, BUT when you have tracks that flow into one another (live albums are the best example) you get a noticeable click. Even if the encoder adds only 0.001 seconds of silence, it is still quite easy to hear the gap. I hope this can be fixed (or already is) as this would be a good advantage over other compression formats. I also wanted to ask how good Ogg Vorbis is in dealing with pre-echo since this troubles almost all encoders, except AAC (Liquid Audio) which does an excellent job in eliminating pre-echo. I know that it is early to ask this question, but a general answer would be appreciated. --- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'vorbis-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.
Do crossover plug-ins alleviate the problem any? --- Aleksandar Dovnikovic <aldov@EUnet.yu> wrote:> [sound gaps in encoded files]__________________________________________________ Do You Yahoo!? Send instant messages & get email alerts with Yahoo! Messenger. http://im.yahoo.com/ --- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'vorbis-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.
I'm no expert. But I was guessing that some of the problems could be from the delay time in loading the song or whatnot. There are pretty advanced crossfades now. Ones that cross automatically when sound levels fall below n decibels, differing degrees of fall-in/out. Variable gap lengths (for fade-in/out) inc. gapless. I'm guessing the "[cross when sound level falls below n decibels]" would remedy your issue. --- Aleksandar Dovnikovic <aldov@EUnet.yu> wrote:> Yes, to a certain degree. Using crossfade plug-ins > you can avoid the click > sound, but that's only a half solution, since the > other song starts playing > before the first one is finished.__________________________________________________ Do You Yahoo!? Send instant messages & get email alerts with Yahoo! Messenger. http://im.yahoo.com/ --- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'vorbis-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.
Vic Forgot (neibwe@yahoo.com) wrote :> I'm no expert. But I was guessing that some of the > problems could be from the delay time in loading the > song or whatnot. There are pretty advanced crossfades > now. Ones that cross automatically when sound levels > fall below n decibels, differing degrees of > fall-in/out. Variable gap lengths (for fade-in/out) > inc. gapless. I'm guessing the "[cross when sound > level falls below n decibels]" would remedy your > issue.You could do all this monkeying or just fix the problem, which is that unknown number of "new" samples are added to the beginning and to the end of the audio material in the encoding-decoding process. I do not know if this actually happens with vorbis , but I know it does with mp3 and it is a major pain in the *ss. The solution could be that the encoder notes the number of extra samples added and then the decoder deletes the same number of samples before giving them out. David Balazic --- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'vorbis-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.
> Anyway, the problem is that all audio encoders out there add a small amount > of silence to the start/end of the encoded file. This is usually a small > amount, around 0.03 seconds. Most of the time it goes unnoticed, BUT when > you have tracks that flow into one another (live albums are the best > example) you get a noticeable click.Vorbis adds no delay or padding to the beginning or end of a sample. Actually, libvorbis right now is missing a feature to make sure the end of a track is not padded. I just recently realized I'd missed that (it's an oversight in our code, not in the format spec).> I also wanted to ask how good Ogg Vorbis is in dealing with pre-echo since > this troubles almost all encoders, except AAC (Liquid Audio) which does an > excellent job in eliminating pre-echo. I know that it is early to ask this > question, but a general answer would be appreciated.Right now, we just use short blocks and can set the trigger threshhold arbitrarily low. This will be hybridized with a partial time-domain pre-coding stage after Vorbis 1.0. Vorbis currently has no real preecho problems (people have accidentally misidentified other bogs as 'preecho'). Monty --- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'vorbis-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.