CCE sound problem

[Luis Neves]

I have two big avi files (3h35m and 1h50m) captured at 480x288 with 44 Khz sound to convert to mpg2 at 480x576 and 48 Khz. I am using avisynth to resize and frameserve to CCE, and CCE to change audio sample rate and encode video and audio. Finally, video and audio streams are authored to DVD with DVD Maestro.

The problem that I have is that sound on the .mpa file after resampled with CCE desapears after a few minutes in both films, and just stays silence. If I use as the source audio not the original .avi file, but a .wav file generated with Virtualdub from the .avi, the .mpa file produced with CCE has the sound ok, reports total size in minutes just exactly like the video stream, but on the film authored with DVD Maestro the sound is out of sync with video!

Any ideias about to solve this problems? How do you usually manage the sound? (p.s. - The original avi plays well video and sound).

[Arky]

Personally, I would bever use CCE to encode my audio. It's fantastic for video, but totally the opposite for audio, IMHO.

I would suggest you use something like TMPGEnc to encode your audio. If it needs upsampling, then SRCC (sic?) would be your best bet for that stage. There is a DanniDin GUI for this malarky as well, I think - I think it's the same one you use for BeSweet encoding).

Sorry I can't be more specific, I'm a little rusty on MPEG1-Layer2 encoding, since I always use AC3 these days. Don't forget that TMPGEnc will accept some audio encoders (TooLame, for example) and upsamplers (including SRCC) as 'plugins', and does a top job. Either Danni Din's GUI, or TMPGEnc will both give you superb results.

Arky ;o)

[Luis Neves]

Hey Arky. I have used TmpgEnc for some time to do this kind of job and I have srcc instaled. My problems started when changed to CCE.

Just finished one test with the shorter film: video was produced with CCE and audio resampled independently with SRCC (through TmpgEnc). Final result after authoring with Maestro: perfect!

So indeed I will forget processing sound with CCE. Thanks for the suggestion.

[Quigonsean]

I would not do sound with cce. But you might try this import the audio that does work but is outta sync into maestro along with your video. Put the audio in audio track 1, then right click in that track and select "Create Audio Sync Track" answer yes to replace current asset and okay to any other pop up. This works wonders, but doesn't do miricals.

Sean

[Luis Neves]

Quigonsean, I have tryed your suggestion and did work fine with sync, but there is a problem; I could only do it with a .wav audio source, and the final size of output has gone from 4.5 Gb to 6.3 Gb. Is there any way to do this with a compressed audio file (mp2 or mpa)?

[Quigonsean]

Yeah, I would download BeSweet, and BeSweet Gui from www.doom9.org in the download section. Convert it to AC3 then do the same in Maestro. BeSweets Gui is very easy to figure out in like 2 minutes. Actually they may have it and a link to its homepage here in the tools section

Sean

[DJRumpy]

Definately. The AssumeFPS command will give you sync'd audio (assuming you used the TRUE flag), but the output is noticably faster/slower, giving you chicks on steroids, or chipmunks on speed

If your not changing the framerate, then the audio from CCE will sound fine, assuing the bitrate is high enough (224 is fine). There are old rumors that the output sounded bad, but a long discussion on it turned out that most just perferred their audio editing elsewhere, or they had simply heard it was bad, and repeated it when asked (they told two friends, and they told two friends, etc). Try it yourself to see. Quality is such a subjective thing.

If you are changing the framerate (say from FILM to PAL), then use BeSweet and the BeSweet GUI as Quigonsean suggested. It has a simple drop down selection for PAL to FILM, and FILM to PAL. Create your new WAV file with the proper framerate length, and use the WAVSource("c:\temp\audio.wav") command in AVISynth (or the AUDIO | WAV AUDIO option) in VirtualDub to feed the resample audio to CCE. Your output will sound fine.

One note: The BeSweet GUI sometimes has a problem if you use the one-click option. In other words, if you give it an MP2 as input, select resample from PAL to FILM, and select MP2/AC3 as output, you sometimes get garbage, or output that isn't resampled. This seems to vary by version. If you experience anything like this, then simply convert your audio to WAV first, then select your WAV file to resample from FILM to PAL (or whatever your converting to). You can input WAV into AVISynth, and VirtualDub, or simply convert it to MP2/AC3 at that point, and mux it back into your MPEG later on in the conversion process.

[Luis Neves]

People, after all the problem was with the avi file, something should be wrong with the sound track. I have made another capture and this time eveything went ok.

So, from the captured AVI I use CCE to encode to mpeg2 the video component, and get an mpv file; with TMPGEnc I extract the audio from the original AVI to an mp2 file (and reencode audio if needed). With DVDMaestro I join again the video and audio and I get a perfectly sincronized result.

Anyway, for any future problem I got the idea about using AC3 and BeSweet. Tx.