Demuxed Delay Help

[Nitro89]

I just ripped Black or White music video from my Michael Jackson Dangerous Short Films DVD using DVDFab HD Decrypter 4. When ripping I had a choice of AC3/5.1 and PCM. I did try AC3/5.1, but when demuxing with dgmpgdec, I did recieve the delay but the audio only used were 2 channels which made the lyrics louder than the instrumentals. I then ripped the PCM, only this time, there was no delay written down, even though when playing back with MPC, there was a huge one. I need a freeware program that can tell me the delay from the .WAV file I received from demuxing the video.

[Nitro89]

I fixed the problem above, but the main problem now is that I now have a .WAV file with no details on its delay, I don't know how to aquire the delay on .WAV files, does anybody know?

[mats.hogberg]

A wav on its own has no delay. It just "is". Delay enters the scene when you add the wav to video - the delay is relative the video, and without the video - no delay.

/Mats

[Nitro89]

so what can I do?

[mats.hogberg]

Add it to the video with zero delay, try to determine how much off sync it is (look for parts with easy to spot audio/video, like a slamming door) and add some silence/cut a little off (depending on if the audio is ahead or behind video) at the beginning of the wav, try again...
IIRC, you can also use MPC, and while playing add/remove A/V delay, that you can then use when you finally mux a+v

/Mats

[Nitro89]

its there a way I can match it exactly to the DVD rip (un-demuxed)?

[mats.hogberg]

I'm not following... Elaborate on that last.

/Mats

[Nitro89]

I need the audio to be perfectly in sync with the video, I was wondering if there's a way I can sync it exactly to the un-demuxed rip of Black or White that I have so I can calculate the delay.

[Nitro89]

When I demux the video, why doesn't the .WAV file like the .AC3 have its delay written on it, the .WAV file is taken from the video

[Midzuki]

Because the .WAV file format was not designed to support such feature.

[mats.hogberg]

What are you trying to accomplish? Why do you demux the a/v, and then mux it back again?

/Mats

[manono]

You might try the same delay as the AC3 track has. No guarantees, though.

[Nitro89]

I tried that, I want to demux it, fixed the delay, remux it, then convert it to MPEG4 for the Samsung YP-P2 player

[mats.hogberg]

Surely whatever you use for the mpeg4 conversion can deal with the source without you demuxing and remuxing first?

/Mats

[poisondeathray]

If it's a simple constant delay problem, just encode it and fix it after

If you use avidemux you can fix it and encode it at the same time.

[Nitro89]

I did have a .VOB file where I did convert it without demuxing it using avidemux but the audio was extremely out of sync.

[poisondeathray]

I'll say it again: if it's a constant delay problem, just fix it (takes a few seconds, maybe 3-4 clicks, no re-encoding.)

If it's "extremely out of sync" as in gets worse as you progress, then you have bigger problems.

I don't understand how 5.1 AC3 is only 2 channels from your opening post?

[Nitro89]

I'll say it again: if it's a constant delay problem, just fix it (takes a few seconds, maybe 3-4 clicks, no re-encoding.)

If it's "extremely out of sync" as in gets worse as you progress, then you have bigger problems.

I don't understand how 5.1 AC3 is only 2 channels from your opening post?

When I demuxed it from the .VOB file with AC3/5.1, it only demuxed 2 channels, the bad left and right which resulted in higher volume in the left speaker and lower volume in the right speaker, I needed to use the option convert AC3 to WAV so it can downmix the AC3/5.1 using all channels to AC3/2 which did work. I did previously try just to demux the AC3/5.1 normally but the delay that was given was wrong.

Also when I rip the DVD, is the audio automatically delayed?

[poisondeathray]

If you ripped the DVD correctly, the audio should be in sync while multiplexed. You can test this by playing it in a software player

Not sure why you're getting all these problems, or why 5.1 audio is being ripped to 2.0 audio.

I'm assuming you're using the lastest version of DVDFab.

Even if you are using the demuxed .wav track, you should be able to fix it "on the fly" adjusting small increments with "audio shift" and playing it back, readjusting in avidemux. Just specify other audio track and select the .wav. Another reason why avidemux should work well, is that it indexes .MPEG/.VOB type files at the very beginning when you open the file. (Much like it creates a VBR time map for .mp3)

If the duration of the .wav and the video do not match (you can check this with mediainfo), then you have a problem. This is the progressively worsening delay scenario I mentioned earlier. Your options are then to either shrink/elongate either the video or audio to match.

[Megahurts]

The 1st VOB file in a titleset has no audio delay. All others have different delays. When playing the Vobs in a media player, the player will adjust for the delay. Use Mediainfo the find the delay.

Load the Vob file in Avidemux. Select Video = Copy, Audio = Copy, check shift and enter the delay from Mediainfo (use (-) for negitive values), Format = MPEG-PS (A+V). The audio track is selected from the Audio menu > Main Track.

The new mpeg file will have no delay.

[Nitro89]

The 1st VOB file in a titleset has no audio delay. All others have different delays. When playing the Vobs in a media player, the player will adjust for the delay. Use Mediainfo the find the delay.

Load the Vob file in Avidemux. Select Video = Copy, Audio = Copy, check shift and enter the delay from Mediainfo (use (-) for negitive values), Format = MPEG-PS (A+V). The audio track is selected from the Audio menu > Main Track.

The new mpeg file will have no delay.

Thank you soo much, your a genius. This means I don't need to demux and remux my VOB's anymore.

[Nitro89]

one question though, in mediainfo, the delay says -345ms but when in avidemux, audio, main track, the shift says -344ms, which is correct?

[mats.hogberg]

Wont matter - you will not be able to hear the difference 1 ms this way or that.

/Mats