Audio Problem using AVISynth and TMPGEnc Xpress

[1980sumthin]

Hi everyone. I have ran into a bit of a problem when following some of the great guides here on this forum about using FitCD to make an AVIsynth avs file to use in my encoder. My encoding Software is TMPGEnc Xpress 4.0 and it seems to work great for me normally, but gives me all kinds of difficulty with avisynth scripts.

What I am trying to do is resize some AVI files to adjust for overscan and convert them into MPEG files ready for DVD authoring. The problem I'm now encountering is that when I load the avisynth file created in FitCD (as per instructions from Fucilives guide), I noticed that the audio goes completely out of sync with the video in TMPGEnc Xpress. If I recall correctly, the video ended before the audio (though I'm not sure if it was the other way around.) In any case it was a good bit out of sync.

What do you think might be the problem and how do I fix it? What's the prober way to convert an AVI file into MPEG DVD via AVISynth and TMPGEnc Xpress?

I have been stumped on this for a very long time so any help on this is so appreciated that I might just have to name my grandbabies after the one who solves this.

[1980sumthin]

Nobody has an idea?

[gadgetguy]

Post your script.

[manono]

What's the prober way to convert an AVI file into MPEG DVD via AVISynth and TMPGEnc Xpress?

I'll assume you mean to say "proper". The usual procedure is to handle the audio outside of the MPEG-2 encoder and bring it in during the authoring stage. In addition to posting your script, you might post your TMPGEnc settings to make sure you're not doing anything foolish like changing the length of the video.

[1980sumthin]

What's the prober way to convert an AVI file into MPEG DVD via AVISynth and TMPGEnc Xpress?

I'll assume you mean to say "proper". The usual procedure is to handle the audio outside of the MPEG-2 encoder and bring it in during the authoring stage. In addition to posting your script, you might post your TMPGEnc settings to make sure you're not doing anything foolish like changing the length of the video.

I wonder if that's the problem. Because I never extracted the audio. But then I would think that even if I did so it would still be out of sync because the script seems to be altering the length of the video during the encoding process. This is what is showing during the filter/edit preview part of the encoding stage (before I even encode, and remains after I do encode). But when I upload the file directly the audio plays in sync perfectly fine. The only problem is then I have to deal with overscan if I were to encode it without avisynth. (I hate overscan with a passion.)

Here's the aviscrypt...

# -= AviSynth v2.5.8.1 script by FitCD v1.2.8 =-
AVISource(the correct location on my hard drive directory)
LanczosResize(656,448,0,2,336,252)
AddBorders(32,16,32,16)
#Trim(0,17983).FadeOut(150)
ConvertToRGB24() # For TMPGEnc or VFAPI

The settings on TMPGenc Xpress are the templates that came in the program. I used "DVD Mpeg" and even tried to use manual mpeg settings using reccomendations I found from searching, and any way I did it, the audio was still out of sync when using avisynth (and only with an avisynth script).

Thank you guys VERY much for you help so far.

[manono]

There's nothing in the script that changes the length. I don't use TMPGEnc and wouldn't know if any of the templates could do so. This out-of-synch, is it constant (off by the same amount all the way through), or is it progressive (gets worse as the video goes along). Constant is easy to fix. Progressive points to a more serious problem.

[gadgetguy]

Your source may be out of synch, with a delay flag that allows it to play in synch, but AviSynth isn't reading properly. If so, this can be adjusted in the script by using DelayAudio(). Try playing your script in MPC and see if the audio is in synch before encoding.

[manono]

That's a good point and one I forgot to mention. It adds to my earlier reply about handling the audio separately. I don't use TMPGEnc but have read that it handles VBR MP3 audio poorly, which can lead to the problem described. Since the audio is still in the script, personally I'd extract it, convert it separately, and remove the audio from the script entirely:

AVISource("C:\Path\To\Movie.avi",False) #False disables the audio
LanczosResize(656,448,0,2,336,252)
AddBorders(32,16,32,16)
ConvertToRGB24()

And, as gadgetguy says, there could already be a delay in the audio that's not being accounted for.

[1980sumthin]

Thank you guys, I will give it a shot.

[1980sumthin]

manano I think you are exactly right about the vbr. (Well first, let me adress the earlier question asked and say that the problem is progressive. It starts in synch, but by 10 minutes or so into it, the audio is lagging behind.) I just opened the file (strait from the source) in VirtualDubMod in order to try your suggestion of extracting the audio seperately, but I got this error...

"VirtualDub has detected an improper audio encoding in the source AVI file (audio stream 1).
The current preference is to rewrite the audio header with standard CBR headers during processing for better compatability. This may introduce up to 830 ms of skew from the video stream. If this is unacceptable, decompress the *entire* audio stream to an uncompressed WAV file and recompress with a constant bitrate encoder. (bitrate 89.2 + 11.2 kbps)

Do you still want to rewrite the header?"

Which is something I really don't understand. What's it saying? What do you think I should do to solve this?

[manono]

What do you think I should do to solve this?

Answer "no" to the question and then go ahead and save out the MP3, either as the original MP3, or decompressed to PCM WAV audio.