-
44.1k audio drifts out of sync on 48k project
Hi guys. Having a common problem trying to sync up a big chunk of 44.1khz audio to a project I want to deliver in 48k (as most of the audio is already 48k). I’ve searched high and low for a solution and have already tried some things.
Firstly I tried resampling the 44.1k audio to 48k with Adobe Audition, but the file it gave me ended up being the exact same length as the old one (and still out of sync with the rest of my audio). I also tried this same method with Vegas, but still no luck.
I’ve tried ctrl+click to stretch by syncing up the start and the end of a clip, but strangely the middle remains out of sync. I was sure this was going to work and I’m baffled that it hasn’t. Now I’m wondering if the 44.1k vs 48k sync is really the issue here, or if the problem is that the 44.1k recording is somehow faster in some parts than others?
Any ideas?
-
5 Replies
-
I have run into this recently with some video I shot a few years ago. In my case, the audio I want to use was recorded by a field recorder (M-Audio), and the cameras shooting the video also recorded the audio. Thus, I can hear an echo when the two sources are out of synch. I am dealing with it by splitting the audio when the echo gets noticeable. I then shift the remaining audio to put it back in synch, and repeat the process. I split the audio in a quiet spot so that the split is not noticeable.
-
Your problem isn’t the 44.1kHz vs. 48kHz sampling. Rather, it’s the difference in sampling clock accuracy.
You tried the obvious solution.. stretch or compress the unsynched track to match. But if that syncs it at the ends but not in the middle, you have a wobbly clock source. This wasn’t digitized from tape or something, was it? What was the source of that track?
-Dave
-
Thanks guys. It was a very old USB recorder so that would make sense. I should be able to cut it up into sections and do my best to sync up each one.
-
I looked up some of the specs of a few older digital recorders… pretty shocking stuff. Without going into too many technical details, it’s all about the accuracy of the clock (quartz crystal, same thing that sets time in just about any modern device). The worst clock crystal I’ve ever used in an electronic device, cheap 100ppm crystals, would give you about a half hour before dropping enough samples to lose sync. That’s bad.
But you can do worse. There are cheaper timekeepers, like ceramic resonators, that might only be 1/5th as accurate. Most cheap digital watches use 25ppm crystals or better. For my last major project, I used 1.5ppm crystals… you could stay in sink for over a day with one of those bad boys. Well, two.. both recorder and camera need to be accurate. But oddly enough, I haven’t seen this problem with cameras, only digital recorders.
-Dave
-
I am particularly disappointed because I was using an M-Audio recorder; a 24/96 microtrak. It is noticeably out of synch with the camera, a Canon GL2, within 10 minutes. I expected a lot more than this from M-Audio. My current Zoom H4N does not have this problem. I also have never seen a problem with a camera.
Reply to this Discussion! Login or Sign Up