Editing AC3 files

When I put a project to DVD, I usually do a Print to Video from my HDV timeline, and then take the tape, and do a firewire output to a set top DVD recorder. I extract the streams with MPEG Streamclip, and reauthor the DVD with DVD Studio Pro.

I want to be able to use the AC3 audio format because my projects can get long and the AIFFs are too big for my DVDs.

The problem is, I want to be able to look at the AC3 waveforms, check them for levels, and perhaps do some basic audio editing (such as changing the volume). I'm looking for a way to edit the AC3 files either in FCP or Soundtrack Pro, but as far as I know, that's not possible.

Is there a good 3rd party app or codec that would work well for this?

Right now I have a PC that runs Nero Wave Editor and I have the AC3 plugin for that, but the program is very buggy and crashes frequently, but when it works, it's a life saver. I'd prefer something for the Mac, though.

Thanks,
Casey Petersen
www.unitedvideoinc.com

Hmm... From what I know, signal processing requires uncompressed pcm signals. You would be better off working with the audio in it's uncompressed form before you send it out to the DVD recorder, else, your .ac3 will end up losing a generation.



www.strypesinpost.com

Let me give you my specific example.

I am editing a wedding. I have separated out the toasts and interviews into separate segments, and there is a 15 minute highlight segment as well. I have put those to tape (HDV), and play them back while recording on my Panasonic DMR-EA18 DVD recorder. I strip out the M2V and AC3 files using MPEG Streamclip, and continue to author my DVD in DVDSP.

In this particular case, the ceremony itself was longer than an hour, so instead of putting it to 2 tapes, and trying (unsuccessfully I predict) to join the two M2V/AC3 clips together on a match frame (impossible since I can only line up within 15 frame accuracy in DVDSP), I chose to export the ceremony using Compressor.

Having done this before where I'm mixing DVD recorder with Compressor, I notice that the audio that Compressor gives me is maybe 5-6 dB lower than what the DVD recorder does, and it sounds ridiculous when I'm watching the DVD to have such a significant audio drop between clips. So what I prefer to do is bring my ac3 file (created from Compressor) into the crappy Nero program and hopefully I can raise the volume to it's proper level and save my new file before it crashes.

BTW, the main reason why I don't compress everything in Compressor is in the interest of time. I am trying to crank these projects out as fast as possible due to tight budgets, and it is a lot faster for me to Print to Video, then do my DVD recorder method than it is for me to use Compressor.

I'm just thinking there's got to be a better way...if a cheap crappy Nero program can do it, why not Soundtrack Pro (or any other audio software that they spent more than an hour programming).

Thanks!
Casey

I agree with you that DVDSP should be able to tweak/modify levels slightly. That can help save a lot of time especially when lining up compilation videos or music in the menus.

The only thing i can think of is to convert it to aiff, tweak it then convert it back. You lose a generation, but that's likely what is happening if you're doing it in Nero.



www.strypesinpost.com

What about exporting the audio from your timeline as a full-quality WAV or AIFF file? The workflow above seems really convoluted.


www.derekmok.com

I can do that, but if my DVD project is already almost 2 hours long, then using full quality audio makes the project larger than a DVD will hold.

What I actually ended up doing in this case, was taking the full quality audio from the 1+ hour long ceremony...for some reason when I tried saving it as an AC3, the volume still went down 2-3 dB. I ended up with a project that was 5.1 gig, so what I did was build the DVD folder, move it to my PC where I have DVD Shrink, and I "shrunk" the 1 hour ceremony at a rate of 90%. It's an acceptable loss of quality (rather minimal actually) because it's not the most important thing on the disc...it's just the ceremony in it's entirety...not my main edit.

A simple codec for Soundtrack Pro shouldn't be that hard, right???

I wish there was an option to gain/degain in DVDSP (not so much quality sometimes, but just to speed things up tremendously).

>What about exporting the audio from your timeline as a full-quality WAV or AIFF file?

I was thinking about that, but since Casey is dubbing it onto tape, and having that go into a set top recorder, i was thinking about the sync....

What I think would be a workflow is to decode that into aiff with compressor, tweak it in Soundtrack Pro, reencode that into .ac3 and he's cool. Not exactly a push of the button workflow, but that should preserve sync, although the dual compression might present a quality hit, so a bit of EQs would help some bits.



www.strypesinpost.com

> I can do that, but if my DVD project is already almost 2 hours long, then using full quality audio
> makes the project larger than a DVD will hold.

Kind of missed the point. I'm saying, keep it full-quality, do the editing, then export the AC3.

As Gerard said, the concern would be sync. Still, the idea of outputting to tape, then dubbing to DVD, then re-extracting the DVD, then trying to use that audio file to do the mixing...it sounds like a lot of unnecessary steps. Are you sure you're saving time here? Besides, what if they're not happy with the mix and want to change it? What if they want to change the picture slightly? Then all your mixing work is kaputt if you do it after the export stage.


www.derekmok.com

Derek: This issue doesn't come up for me very often...maybe 3-5 times per year (and it's a hassle when it does come up), but these cases, there's really not much demand for a remix in the audio track...I think the only things I've had to change in wedding DVDs over the past few years are just spelling errors (either mine or theirs), and once I had to remove a guest out of the highlight video.

Gerard: I think I'd still have the same volume issue, though. I think Compressor is the thing that's bringing the volume down...shouldn't be though, right?

I'm able to make this work the way I'm doing it...I haven't been totally screwed yet, although 1 of my 2 PCs won't run Nero...it crashes when I try to open it.

I suppose I could try for more consistency and use Compressor all the time, but if I finish editing a project at 3:00, I can still get it done by 5:00 by using the DVD recorder, whereas I'm looking at encoding from 3:00 till night time, well after I'm gone, and finishing the DVD in the morning.

Again, this is more of an exception because of the ceremony being longer than a single tape will hold. I could have used a longer tape, but I chose not to drive 40 miles round trip to get one.

Thanks for your help!

Casey

You don't need to use compressor though. You can just get DVDSP to do it in the background while you work on other things. I think you're doing it the long way round. Just drop it into DVDSP and author it as you go, with DVDSP set to encode as it goes.

When you encode as AC3, set dialnorm to -31. You're going with the Dolby default which sets it to -27. I could do a long, involved explanation of why this is the way to go (that would be boring!), but fixing the dialnorm parameter will work to avoid that volume loss you're hearing.

>I think I'd still have the same volume issue, though. I think Compressor is the thing
>that's bringing the volume down...shouldn't be though, right?

Yup. That's dialnorm. -31 toggles dialnorm off. Also check if "film standard compression" is turned off a few tabs down.



www.strypesinpost.com

Never heard of it. I'll give it a try next time

Jude: I'm working with HDV, I can't drop my QT movie onto DVDSP in a standard def. DVD, can I?

KK Proffitt: I'd be interested in hearing a little more about what a dialnorm is...it doesn't have to be very long...just the basic principle of the thing.

Thanks!!!
Casey

Quote
The Dialnorm scale spans -1 to -31dBFS in 1dB increments. The -31 setting represents no level shift, while -1 is the maximum level change. Dolby Digital standardises the time-averaged loudness of dialogue to -31dBFS by applying a fixed level-shift to all 5.1 channels determined by the dialnorm setting. If the dialnorm setting is -31, for example, no level change is required, as the programme already conforms to the required average level. On the other hand, if the programme has a more compressed dynamic range, it may be given a dialnorm value of -21, to reflect the fact that the average dialogue level is closer to the peak level. In this case, the decoder will apply 10dB of attenuation to bring the average level of the programme dialogue back down to the standard level.

from [www.soundonsound.com]



www.strypesinpost.com

And the reason you can't change anything after Ac3 encoding, including volume, is that the encoding of the file embeds metadata into the file itself. Changing anything after the fact would render at least some of that metadata meaningless.

Reid C

I think Strypes beat me to it! Be aware, though, that even though you turn off all compression, some consumer players will ignore your setting and compress it with their default setting anyway. It can be maddening.

The quote from Strypes is really about the center channel, or where the dialogue lives most of the time in 5.1 film and video audio. It's typically measured with an LM100, a device from Dolby. Unfortunately, dialnorm affects the total level, even if you only have two channels, so it's been the subject of many a heated thread in music forums.

Also there's an internal limiter in the AC3 encoder that checks to see if your total volume is too loud for the encoder. If it doesn't pass muster, the audio gets hit hard with a limiter and sounds awful. I hardly ever allow peaks above -6 dB (but I'm mixing in a Dolby cal'd room, so it pretty much takes care of itself).

Here's one I found that will convert AC3 files to AIFF or whatever and back again.

SoundConverter ... and it's only $14!

Casey Petersen
www.unitedvideoinc.com