Times when I'm so happy I have Vegas!

PeterWright wrote on 9/12/2012, 10:52 PM

I recently shot a boomerang demonstration, with the main talent wearing a wireless lapel, but there was still lots of wind and traffic noise. I typed an exact transcript of everything he had said and got him to record the audio again.

He did it well, but just about every phrase, and sometimes an individual syllable, had to be split then sped up or slowed down to match the original wave form. Ctrl/Drag with Elastique in Vegas makes this easy - it took the best part of a day's work to re-synch a 7 minute clip, and apart from an occasional artefact from excessive stretching, I reckon you'd be hard pressed to spot that the audio is not original ..... except that I've told you! The result was of course "too clean" so I made up a clip of quieter ambience from the location and added that for "reality".

The clip is here:

http://www.boomerangsrus.com/videos.htm

Comments

i am erikd wrote on 9/12/2012, 11:51 PM

Impressive indeed! Good work.

farss wrote on 9/13/2012, 4:24 AM

That's a pretty decent effort, I've tried doing much the same and given up as it was impossible to avoid breaking the cadence of natural speech. In some places this is pretty obvious to me in that piece but I did spend a lot of time working with dialog for talking books etc.

I would think using ADR would have overall involved less work and it is pretty simple to do with Vegas. If that wasn't feasible I wouldn't have tried to completely match the original waveforms, some loss of lip sync in those scenes wouldn't be all that objectionable.

Bob.

Tim20 wrote on 9/13/2012, 5:11 AM

Very good. I doubt most people would notice. But I can certainly hear the artifacts and immediately noticed his recorded voice didn't match the acoustic space he was in. i.e. outside. Outside has little to no reverb, but I can hear the reverb from the room it was redone in.

PeterWright wrote on 9/13/2012, 5:12 AM

Bob - I had overlooked ADR, and have been reading up on it since your post.

I think you're right - it would have taken less time, in fact I'll definitely give this a try next opportunity, but I did enjoy employing one of Vegas's many talents.

Have I got this right? The talent listens to themselves through headphones, from the Vegas timeline in preselected "chunks", and when they are ready, they join in and repeat the performance and we record them onto a new track. The thing in Vegas which I think you mean makes it simple, is that when recording in a preselected Loop (the length of the chosen "chunk"), Vegas automatically records each loop as a Take, and we can access every one of these later on the Timeline by toggling "T".

I think I'd still go to the trouble of typing a transcript to help the re-recording process.

PeterWright wrote on 9/13/2012, 5:15 AM

Thanks Tim - I did try and reduce that reverb, but obviously not enough!

Tim20 wrote on 9/13/2012, 5:22 AM

Yes you are partly describing punch in recording. However if you don't have a vocal booth you need to match the recording space of the orignal otherwise it will sound very different due to the room acoustics. That reverb thing I was talking about. With a booth which is dead sounding you can apply reverb to match the original. An art in and of itself.

You can partially eliminate reverb by using EQ to take out some of the high end.

farss wrote on 9/13/2012, 6:24 AM

"Have I got this right?"

Pretty much. I gave the talent a mix from the original plus their mic as well as a "control room" mic so I could talk to them. Having the vision for them to watch at the same time helped as well of course. It did take time, I think 20 minutes for one line but the talent wanted to get it right and it was a difficult scene where he has the girl against a wall just after the've come into the house and are about to do the deed.

All the kit I had was an uber expensive Sanken shotgun to kill all the room tone, which was my loungeroom, M-Audio Firewire 410 and some old version of Vegas.

Yes, you're right, just leave it looping and Vegas records takes, rolling the shot before and after also helped the talent.

In the final mix, watching it twice in a big preview theatre I was blown away that no one noticed my ADR work, so well was it done I had to fight for a credit for it :)

Sure having a script is important but don't let the talent try to read it aloud, that can make it impossible for them. Give them enough space to "act out" the scene or shot rather than sitting or standing dead still. In your case nothing is uber critical, it isn't a narrative and all you need is clean dialog that isn't distracting in its nature and at the same time shouldn't sound like a perfect VO because it will not match.

Bob.