Why can’t PluralEyes sync camera footage with an audio recorder?

[Dave Haynie]

I do this all the time. I generally use one digital audio recorder as the effective timebase, since it will be run through the entire shoot, while I may be starting and stopping cameras. No obvious reason it shouldn’t work the other way.

The audio/video sync shouldn’t be any worse than two camcorders, unless you have a poor audio device (early Zooms were good for less than 20 minutes, typically, but they fixed that in the H4n and other recent units). One hour might get you a slight de-sync, but I’d deal with that in a second pass.

Of course, make sure your video segments are split by time on capture. A single clip that’s got a time discontinuity will be unsyncable. I haven’t used the “events in order” switch. Not sure what else could go wrong, it has always pretty much just worked for me.

-Dave

[Colin Morris]

Hi John,
I have one suggestion-have you tried inserting a dummy video track between the two audio tracks? It worked for me with the trial version awhile back. The other option is to use Woowave Synch Pro. I have a few friends who use it and they really love it.
https://woowave.com/

[Nigel O’neill]

Curious. Can’t comment on PluralEyes as I sync by eye and ear, but like you, I usually have continuously recording events. I have not had a sync issue with any of the long events I have recorded over 80 minutes, and I am mixing tape (FX1) with data recorder (Z1P + Datavideo DN60), memory card (PJ760), with an XLR feed going into an NX5. I have had some sound delay issues with the audio feed from the sound desk as their compression tends to add a delay, but once the audio is lined up, it stays that way. It does not drift.

There is a suggestion I have read on other forum of setting the Marantz PMD-660 to record at 99.9% speed to match the 29.97 fps of NTSC.

[Igor Jovcevski]

Hi there. I am the WooWave developer and would invite all to try out woowave. If you can’t afford it, you can get a license just by simply posting your process on youtube(screen record) . All you need to do is compare woowave to other app doing similar thing and post the video online(comment would be great but not essential) What is important though is to provide link to the footage used, or proxy of the footage so anyone can recreate the tests. Enough of biased reviews and PR. Woowave will always post footage used in tests. Fair& Transparent 🙂

Igor
Woowave.com
tools for digital artists

[Geoff Candy]

John/

Just a long shot. PluralEyes comes as a standalone as well as plugin. Maybe the standalone version would work. I’ve tried a short piece shot with A1E Vid+Audio with independant picked up using a Zoom. Just followed a tutorial and it worked fine. Not tried it from the timeline yet.

[Igor Jovcevski]

It all depends on the source audio. In my own case, when I developed WooWave, I had to do hundreds of tests with different audio. Tradeoffs are usually unusual frequencies. without having access to some of you audio, one can’t really say much 🙂 . In my own case, the biggest problem was to get rid of false positives or get less.
What I do in WooWave, and what is probably done in Plural Eyes is applying Fourier transform with window function to get Spectrogram of the audio. Then this ‘image’ is analyzed for patterns in all audio clips. If they match, sync is established as an offset. In this case, if it works with standalone, xml parser is to blame, but I would guess that the problem lies with the nature of the source audio . The above is oversimplified description 🙂 There is so much to this to get it going… lots of small tricks and innovative algorithms to get precision and speed.

Igor
Woowave.com
tools for digital artists

[John Rofrano]

I plan on trying woowave when I get home tonight (assuming there is a trial to try). I’ll let you know how I make out.

My guess is that the reverberation from the auditorium probably produces a significantly different spectrogram from the FFT than the dry audio coming from the desk which is causing sync to not be established. I’ll load both up into Sound Forge and look at the spectrograms myself. Still, you can see that the waveforms look similar in Vegas because that’s how I’m lining them up by eye and have been for years (simply using amplitude).

We shall see.

~jr

http://www.johnrofrano.com
http://www.vasst.com

[John Rofrano]

[Nigel O'Neill] “There is a suggestion I have read on other forum of setting the Marantz PMD-660 to record at 99.9% speed to match the 29.97 fps of NTSC.”

That’s an interesting concept. I still can’t see how you can run for 80 minutes without drift and no genlock. The crystals in these devices simply aren’t that accurate and yet I believe you if you say it’s working for you.

~jr

http://www.johnrofrano.com
http://www.vasst.com

[Igor Jovcevski]

There is a trial 🙂 . Not only that, but if you post screen recording of your session while comparing WooWave and Plural Eyes on the same footage an dpost that online, you get a license for free. woowave.com/woowave-challenge There is only one requirement, footage used must be available so others can recreate the test. If the audio is very long, woowave may take much longer to process, but it will eventually find the sync (especially if you choose ‘sacrifice precision’ option before starting sync) This option is best for noisy voice recordings. I would like to see how you’re getting on with it. all suggestions are precious 🙂

Igor
Woowave.com
tools for digital artists

[Igor Jovcevski]

One more thing about WooWave.. it tolerates quarz imperfections for up to 32ms.

Igor
Woowave.com
tools for digital artists