Good tip for syncing audio tracks.

[Hulk]

While editing a video just now and struggling to line up two audio tracks I remembered trick I used to use for syncing samples with actual sounds.

Put the pan audio filter on one track and pan it hard left for the duration.

Put the pan audio filter on the other track and pan it hard right for the duration.

Put your head between the speakers and see which way the sound seems to be moving when you hear the front edge of a sound. If it's going right to left then the right track is ahead a bit.

It's pretty easy to get tracks spot on this way once you get good at it.

I'm actually still editing my wedding and I'm putting the actual entrance song behind the one through the PA on the tape, I can still hear the announcer but now the background track it nice and close, also, I can duck the track when I want to hear the announcements.

Mark

[Brian Ellis]

I use the waveform mode of MSP. You can easily synch it to about 1/20th of a frame. However, what I have noticed is that it also depends on the distance from the sound source. Imagine that you are filming a cricket or baseball match with two cameras, one as close to the players as you can get to get the action and the other farther away to get an overall view. The batsman swipes the ball with an almighty crack. If you synchronise the two tracks on the sound, the motion will not be smooth if you cut from the close-up to the overall view, because the sound has different distances to travel. In fact, it is much better to split the sound and slide it along so that the crack happens as the ball hits the bat for both tracks (technique much used in pro work: ever wonder how you hear the explosion as you see it, even if filmed from kilometres away?). I don't know whether this would be important in a wedding in a large church/synagogue/mosque/temple/whatever. I do remember, sitting in the back of the Mormon Tabernacle in Salt Lake City during the famous pin-dropping demo of the acoustics, that I did notice a slight delay between sight and sound, but I doubt whether this would upset the lip-synch of the priest/minister/rabbi/imam/lama/whatever or the marriage victims, because, if the camera were that far away, the lips would be hardly visible.

On one occasion, in order to get decent music quality to dub in at a pro classical guitar concert in a hall with terrible acoustics, in the days when using VHS-C and because the real aim was to produce sound cassettes, I did the recording on a Nagra at 15"/sec with a single cardioid condenser mike positioned just 50 cm from the guitar body, on the artist's left side, so that any fingering noises were minimised by the mike response. This was before I did NLE and, with great difficulty, I was able to dub in the good sound onto the video while linear editing. Even though the two sources were asynchonous, I was able to maintain good synch for up to about 1-1/2 to 2 minutes at a time and I was able to restore it during a brief transition to the audience. It took me about 5 days to do this for a 90 minute concert! Unfortunately, I lent the tape to someone and never got it back. I still have the original Nagra master tape, but no Nagra to play it on! (At the time, I worked for Kudelski, the manufacturer of the Nagra, which was why I had access to the most fantastic pro sound equipment in the world.) . This made me think that if I had had MSP and DV in those primitive days, it would have been a far easier task. I venture to think that I could have kept synch over at least five minutes.

[jeff b]

I taped a music performance last year using two cameras plus an audio tape made off the console on the stage. I got in view of both cameras and near enough to one of the microphones running through the console to get picked up and clapped my hands. It made synchronization in post very simple.

Jeff B

[Hulk]

I have A LOT of experience with studio recording so I know what you guys are talking about. BTW Brian, I like to record acoustic guitars with a close small diaphram condenser mic, about a foot or so from the hole and a large diaphragm condenser (414 is nice) about 5 or 6 feet away. Remember, this type of miking would be heresy for recording classical or jazz but I was going to the interpretation of the artist, or what was in his/her head, not so much the reality of the room. Middle side miking with a 414 on figure 8 pattern with the output split into two signals, one signal panned hard left, the other phase-inverted panned hard right, with a small condenser on top of the 414 to maintain phase coherence with center mic panned straight up the middle creates a nice acoustic guitar sound as well, plus you can focus or widen the sound by adjusting the panned and center mic levels.

Anyway, using a zoom on the audio waveform would not work in this case because I was syncing a song played through the dj's pa system with the actual track. Obviously there is so much noise in the camera audio track with audience, room etc that the zoomed waveforms didn't even look like one another. In this case panning each track hard right and left and listening was really the only option.

It worked out really well. As the DJ would announce the bridal party members I would duck the song audio track and then bring it back up when he stopped talking.