The No.1 Website for Pro Audio
 All  This Thread  Reviews  Gear Database  Gear for sale     Latest  Trending
Aligning audio with live video?
Old 16th October 2013
  #1
Lives for gear
 
GZsound's Avatar
Aligning audio with live video?

I am going to be doing a multitrack recording and video of a concert in a couple of weeks. I don't do anything with video, but my wife does and is all over You tube but with camera sound.

Can someone suggest a simple way to mark both video and audio for alignment later? We don't have the capability of locking time code, etc.

Would a simple video and audio mark of say clap my hands on stage work? Any suggestions would be appreciated.

We want a video with professional sound and I can remix the tracks in my studio and get a pretty good product.
Old 16th October 2013
  #2
Gear Addict
On simpler low budget projects I have used the hand clap method. Be sure to have a very clear image of the hands. An actual clapper is better.

Danny
Old 16th October 2013
  #3
Lives for gear
A clapper board never fails,visual and sonic cues,the best metadata yet.
Does require someone who has a clue though and keeps the board up to date (ie a camera asst.)
Old 16th October 2013
  #4
Lives for gear
 
hbphotoav's Avatar
 

Couple of things... If all the cameras are close to the stage, and all are good for the same length of take, clap synch is a good start. If some or all are DSLRs with limited takes (usually 12-30 minutes) and the program is long-form, and the cams are restarted at different times, or are mixed with long-form AtomOS or AJA disk recorders (or, heaven forbid, tape) a waveform analysis/synch program like PluralEyes (or physically matching waveforms within your edit program) will be necessary.

If you are using PluralEyes or a similar program for aligning waveforms from the discrete cameras' onboard "scratch" audio, and are at varying distance from the stage, the synch between camera "scratch" audio and the board/tracking mix will need to be adjusted about a frame for every 35 feet or so that the camera's mics are from the stage.

HTH.

HB
Old 16th October 2013
  #5
Lives for gear
 
Richard Crowley's Avatar
 

My "big rig" was hundreds of miles away doing another production, so I did a simple document of this choral conducting workshop with a small "handy-cam". Alas, the audio in the camera went out early in the concert, and I had to sync most of the songs purely on visual cues. In this example, there were so many drum hits at the beginning, I was unable to establish a decent sync, so I had to go to around 0:59 and sync on the "L" in "new Jersusalem". (Carly Simon's "Let the River Run")

https://vimeo.com/71927519

(The Vimeo tag on this forum doesn't seem to work at the moment)

There are several ways to sync audio and video. It is almost never a good idea to "tweak" the audio track to match the video. You open the possibility (probability?) of making a hash of the audio track by trying to change length. Especially when you then data-compress the audio into a delivery codec. IME, it is much more accepted by the audience to "tweak" the video to match the audio track (which I use as the "master"). A duplicated (or dropped) frame every few seconds is not nearly as noticeable as a metallic tinge in the audio from time-stretching. And it is trivial to "pull-up" the video to match the master audio track whenever an edit point is inserted.
Old 16th October 2013
  #6
Lives for gear
 

I use Sony Vegas. I just visually align the waveforms. I.e., I slide the external audio until the peaks line up with the camera audio. If things get out of synch I put in a transition. Oh, and if your cameras are far from stage there can be a time delay (speed of sound and all).
Old 16th October 2013
  #7
Lives for gear
 

Quote:
Originally Posted by GZsound View Post
I am going to be doing a multitrack recording and video of a concert in a couple of weeks. I don't do anything with video, but my wife does and is all over You tube but with camera sound.

Can someone suggest a simple way to mark both video and audio for alignment later? We don't have the capability of locking time code, etc.

Would a simple video and audio mark of say clap my hands on stage work? Any suggestions would be appreciated.

We want a video with professional sound and I can remix the tracks in my studio and get a pretty good product.
With longish concerts shot with DSLRs (that auto-cut @ 20 min or so) the best we've done is to make sure each camera has a solid reference audio feed--in our case a mono mix on a wireless. That gives you a solid base for autosyncing cam audio to master audio in PluralEyes or FCX etc. The sync should be fine if the clips are that short. If you are shooting cameras that can roll for much longer--say for a whole hour+ set, you will almost certainly have sync drift between the cameras and between the cameras and audio. That you will fix in editorial. As was pointed out above--even numerically correct sync starts to look wrong the farther away from the action the camera is. We have a saying: "Sync is in the eye of the beholder"--meaning that you'll be adjusting the sync of shots anyhow to get them to look right, so just getting the sync close over a long take on location is usuallty good enough for something that will have a lot of editing done.

philp
Old 16th October 2013
  #8
I import the video into Pro Tools; do a rough placement of the audio on the timeline; scrub to find the onset of an obvious note; trim to that point; then, zoomed all the way in on the audio, grab it and (watching the video window) drag it until the attack matches the visual of the action that caused it.

Restore the rest of the audio to the beginning with the trim tool and voila - instant hand sync.

Takes about 2 minutes (not counting import times) and most times is far more accurate than PluralEyes; works with or without reference audio.

A clap sync is nice to have when possible, but often I don't have the luxury.
Old 17th October 2013
  #9
Gear Guru
 

a handclap gets you close

aligning the recording to the camera mic is usually how I fine tune.

if the camera is farther away than it appears, like if it is zoomed in, some tweaking may be necessary as there is an expectation of delay with distance.

Quote:
Originally Posted by philper
We have a saying: "Sync is in the eye of the beholder"--meaning that you'll be adjusting the sync of shots anyhow to get them to look right,
this
if it looks wrong, it is wrong


you can get electronic clappers and there are even apps for the iPod
Old 17th October 2013
  #10
Lives for gear
 
GZsound's Avatar
Thanks very much for the suggestions and the help guys, I really appreciate it.

What we are doing is a single simple video using a Canon digital camera..your basic $300 hand held.

I am running sound for the concert so I am going to use my Cymatic LR16 and do a multitrack recording off the board with an audience mic. I plan on starting and stopping the recording between songs to avoid having one huge multitrack file to deal with.

My wife uses Sony Vegas to edit all her video for YouTube and it appears it is fairly simple to line up the audio with the video. If I can get one of the band members to clap once right before beginning a song it might be easier. Maybe even a down stroke chord on the mandolin would work.

Thanks again..
Old 17th October 2013
  #11
Lives for gear
Have them use the Lawrence Welk introduction of giving the name of the tune followed by "one and a two and a three and a ...."
Old 17th October 2013
  #12
Lives for gear
 
hbphotoav's Avatar
 

Mark... were it me, I'd let the concert run and edit the mixed down track as needed to compensate for any drift of the camera's time basis against the audio recording. Snipping or adding a few frames of audio through a crossfade (especially during applause or room-tone breaks between selections) seems easier than dealing with more than several multitrack clips in the same fashion.

Let us know what you decide and how it works...

HB
Old 17th October 2013
  #13
Lives for gear
 

Quote:
Originally Posted by hbphotoav View Post
Mark... were it me, I'd let the concert run and edit the mixed down track as needed to compensate for any drift of the camera's time basis against the audio recording.
I agree. It's also easy to forget to stop the recording, or worse, to forget to start it back up again. Much better to just let it run. You can always chop it up later if necessary.
Old 17th October 2013
  #14
Lives for gear
Syncing is not a situation of "the more [files], the merrier." One audio file and one camcorder file is enough fun.

If the Mrs. doesn't video between songs to save battery power, you might consider getting an extended battery pack if camcorder battery life is a concern.

Close up video is easier to pick out points to sync than distant shots.

I used the wrong analogy for bluegrass when I brought up Lawrence Welk. I should have said the Darlings from the Andy Griffith show counting off "1 ....2...."
Old 18th October 2013
  #15
Lives for gear
 
GZsound's Avatar
Quote:
Originally Posted by hbphotoav View Post
Mark... were it me, I'd let the concert run and edit the mixed down track as needed to compensate for any drift of the camera's time basis against the audio recording. Snipping or adding a few frames of audio through a crossfade (especially during applause or room-tone breaks between selections) seems easier than dealing with more than several multitrack clips in the same fashion.

Let us know what you decide and how it works...

HB
That's more or less what I plan on doing. I am going to let the camera run and just stop and start the recorder. Again, that's mostly so I have each song in an individual folder. I agree we can do camera crossfades, etc. to make up for any timing difference between audio and video.

I am going to do a dry run recording at the same concert venue this Saturday to make sure the recording works well off the FOH mixer and an audience mic also works. The real deal is next month.

Thanks again.
Old 18th October 2013
  #16
Lives for gear
 
GZsound's Avatar
Quote:
Originally Posted by 2manyrocks View Post
Syncing is not a situation of "the more [files], the merrier." One audio file and one camcorder file is enough fun.

If the Mrs. doesn't video between songs to save battery power, you might consider getting an extended battery pack if camcorder battery life is a concern.

Close up video is easier to pick out points to sync than distant shots.

I used the wrong analogy for bluegrass when I brought up Lawrence Welk. I should have said the Darlings from the Andy Griffith show counting off "1 ....2...."
The Mrs. will be on stage playing. It's her bluegrass band. I am going to let the camera run. We have it connected to it's power cord so battery life isn't an issue, nor is SD card capacity. She's only doing two sets. Gonna be interesting. If it works, you'll all be notified of the Youtube location.

Thanks again everyone.
Old 20th October 2013
  #17
Gear Maniac
 

If the camera is stationary, feed audio from the board directly to the camera. That will make the manual syncing that much easier. You can use the zoomed waveforms to visually sync the audio (the direct sound feed eliminates the delay of the sound traveling slower than light as the action crosses the room to the camera).

Personally, I would let both the camera and the audio recorder roll, if they are capable. Actually, I'd try to borrow a second camera and shoot some b-roll (feed audio if that camera is also far from the stage). Even if the two recordings drift, it's much easier to trim/insert a few frames of video here and there to keep everything flowing than it is to try to sync up a bunch of clips- yes, I've done it...!
Old 20th October 2013
  #18
Lives for gear
 
GZsound's Avatar
Quote:
Originally Posted by justDoug View Post
If the camera is stationary, feed audio from the board directly to the camera. That will make the manual syncing that much easier. You can use the zoomed waveforms to visually sync the audio (the direct sound feed eliminates the delay of the sound traveling slower than light as the action crosses the room to the camera).

Personally, I would let both the camera and the audio recorder roll, if they are capable. Actually, I'd try to borrow a second camera and shoot some b-roll (feed audio if that camera is also far from the stage). Even if the two recordings drift, it's much easier to trim/insert a few frames of video here and there to keep everything flowing than it is to try to sync up a bunch of clips- yes, I've done it...!
Thanks. I might go ahead and do as you suggest, run audio to the camera from the board.

However, as to letting the recording roll and then sync to video, you are thinking of a stereo recording. This is a 16 channel multitrack that I need to edit, mix and master. There is no way I could do that to two hours of multitrack recording in one large file. For example, any cut or timing shift in song two would translate to a timing shift down stream for that track in the next twenty songs. An ugly proposition.

We are aiming for a nice video clip with the best possible sound so it has to be done song by song. Having an audio stream to the camera makes a lot of sense.

thanks...
Old 20th October 2013
  #19
Lives for gear
Not all consumer camcorder preamp can handle a soundboard feed.
Old 20th October 2013
  #20
Lives for gear
 
GZsound's Avatar
Quote:
Originally Posted by 2manyrocks View Post
Not all consumer camcorder preamp can handle a soundboard feed.
Our little Canon can. Plus, I have gain control on the stereo output. I give a board feed to many bands at that concert location and thus far nobody has claimed the sound feed was overdriven.
Old 20th October 2013
  #21
Gear Nut
 
timothyclee's Avatar
 

Final cut x does this automatically. If you are doing it by hand just drag the separate audio file into your video editing program and play both tracks together. Move the audio forward or back until it both I'd syncs together. Be aware most programs only let you move in frame increments.


Tim
Old 20th October 2013
  #22
Gear Guru
 

Quote:
Originally Posted by timothyclee View Post
Final cut x does this automatically. If you are doing it by hand just drag the separate audio file into your video editing program and play both tracks together. Move the audio forward or back until it both I'd syncs together. Be aware most programs only let you move in frame increments.


Tim
In Pro Tools, you can nudge the audio against the video by as little as one sample. The video of course can only be nudged by the frame. One-frame is just too coarse for me.

Your video needs to be 'finished' before you bring it in. That is to say PT can only import one video file at a time.
Old 21st October 2013
  #23
Gear Maniac
 

My advice was based upon the precept that everything could just be left recording for the duration. I multi-track live performances almost exclusively, with sets ranging from an hour to more than four hours on quite a few occasions. With continuous recording, it is easy to sync everything once, then tweak each source as needed to compensate for drift.

If you take the "individual song" approach, you'll want to slate each shot, which didn't seem like an option given that you are capturing a performance.

When working a live performance, I strive to remain invisible...!
Old 21st October 2013
  #24
Lives for gear
 
GZsound's Avatar
Quote:
Originally Posted by justDoug View Post
My advice was based upon the precept that everything could just be left recording for the duration. I multi-track live performances almost exclusively, with sets ranging from an hour to more than four hours on quite a few occasions. With continuous recording, it is easy to sync everything once, then tweak each source as needed to compensate for drift.

If you take the "individual song" approach, you'll want to slate each shot, which didn't seem like an option given that you are capturing a performance.

When working a live performance, I strive to remain invisible...!
I like your idea and appreciate the advice. However, do you have to auto tune vocals? Line up harmony parts? Replace bad chords or notes? Take out pick clicks, foot taps, etc. per song? I may have to do that within any individual song and trying to do that with a continuous file would be a nightmare. We are striving for the very best audio possible under each video clip.

I am going to figure out a way to "slate" each song. Maybe have the mandolin player just hit a quick chord right before the song starts. Better than a hand clap.

Again, I'm glad I asked the question here, you all have given me some valuable ideas.

Thanks again.
Old 21st October 2013
  #25
Gear Maniac
 

Not sure I understand. Why is one long session a problem? I've been recording shows this way for years. I'll start off mixing the whole show, getting my desired plugins in place and getting the overall sound together, then chop it up into individual songs for fixing individual problems. This gives the collection some cohesion when played back as a whole.
Old 21st October 2013
  #26
Gear Maniac
 

Record it all then export the songs separately after, it's what I do for long and large channel count multitracks. Then sync to a scratch audio bounce w/ full length, so that after you've mixed the separate songs you have a nice waveform cue, also a nicer audio to edit the video to while the finished mix isn't ready.

Sent from my mobile device
Old 21st October 2013
  #27
Lives for gear
 
Richard Crowley's Avatar
 

Quote:
Originally Posted by GZsound View Post
.., do you have to auto tune vocals? Line up harmony parts? Replace bad chords or notes? Take out pick clicks, foot taps, etc. per song? I may have to do that within any individual song and trying to do that with a continuous file would be a nightmare.
And that is why we break up the live tracks into individual songs to work on.

Quote:
We are striving for the very best audio possible under each video clip.
And that is why starting and stopping the audio tracking is a terrible idea IMHO. It takes an unpredictable amount of time to write the buffer and close the files, and it is VERY EASY to miss the next song because of that unpredictable overhead. I would NEVER attempt to do such a risky thing. And one of the downsides of the Cymatic is that (at least judging by the published specs) the file close operation seems to be rather on the slow side.

Quote:
I am going to figure out a way to "slate" each song.
Why? Surely you can tell which song is which by listening. You may be in danger of over-thinking this. Record a continuous multi-track of the entire set, break it up into songs, do all your post processing, and then use the mixed-down stereo track as the master to pull-up to video to sync with.

I do this rather often and don't find it nearly as difficult as you are making it out to be.
Old 21st October 2013
  #28
Lives for gear
 
GZsound's Avatar
Quote:
Originally Posted by justDoug View Post
Not sure I understand. Why is one long session a problem? I've been recording shows this way for years. I'll start off mixing the whole show, getting my desired plugins in place and getting the overall sound together, then chop it up into individual songs for fixing individual problems. This gives the collection some cohesion when played back as a whole.
O.K. now I understand. Eventually you chop it into individual tracks to fix individual problems. I just jump to that step first.

With my computer, if I do any edits on a 16 track hour long file, it takes forever to render the edit. Much faster if the tracks are separated into song tracks first. Since I only use plugins on final mastering I would only need them on a per song basis anyway. I can't imagine a "global" plug in that would work for every track in every song.

You must have a lot more horsepower than I have available.
Old 21st October 2013
  #29
Lives for gear
 
GZsound's Avatar
Quote:
Originally Posted by Richard Crowley View Post
And that is why we break up the live tracks into individual songs to work on.


And that is why starting and stopping the audio tracking is a terrible idea IMHO. It takes an unpredictable amount of time to write the buffer and close the files, and it is VERY EASY to miss the next song because of that unpredictable overhead. I would NEVER attempt to do such a risky thing. And one of the downsides of the Cymatic is that (at least judging by the published specs) the file close operation seems to be rather on the slow side.


Why? Surely you can tell which song is which by listening. You may be in danger of over-thinking this. Record a continuous multi-track of the entire set, break it up into songs, do all your post processing, and then use the mixed-down stereo track as the master to pull-up to video to sync with.

I do this rather often and don't find it nearly as difficult as you are making it out to be.
You may be right. However, I have used my Cymatic several times now and the time between stop/write and start/record is less than 15 seconds. But I am beginning to agree with you and will most likely record one file per set and break it up later.

thanks..
Old 21st October 2013
  #30
Lives for gear
 
Richard Crowley's Avatar
 

Quote:
Originally Posted by GZsound View Post
I have used my Cymatic several times now and the time between stop/write and start/record is less than 15 seconds.
For a recording of what length? The longer the recording, the longer it takes to close the file and get ready to record again. IME, 15 seconds is FOREVER when you are trying not to miss anything.
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump