Voice-overs?

[Greg Boston]

Originally Posted by Luke Loeffler

Does anyone know if there are cameras with XLR ins and a firewire connection that could be configured as a real-time audio input device in lieu of the many dedicated firewire/USB mic/line interface boxes available?

Yes, I used my XL2's XLR inputs and fed the firewire output into a laptop running Scenalyzer in audio capture only mode. Took the resultant .WAV files and fed them into Audacity for rough editing. Eventually loaded those rough edited files into Soundtrack Pro for final sweetening and dialog replacement.

-gb-

[Paul Cypert]

sounds like a messy, time-consuming process...glad it worked for you...I've found some nice firewire input devices offering up to two mic inputs for around 120 or so...might check one of them out...some that even allow you to stack and work with bigger inputs the company also offers if you want to upgrade down the line...plus the maudio one now supports Pro Tools...that's a nice cheap way into the Pro Tools line up....anyway, I'm getting excited...don't know how in the world I'm supposed to record a VO track in Laotian :)...I know a little Thai which is close, but this is going to be quite the process and I'm going to have to trust my Voice Talent quite a bit...

Paul

[Matt Davis]

Originally Posted by Paul Cypert

at these prices I'd just assume get another Mbox....any 100 or sub input devices out there?

This may sound cheap and miserley, but if you connect your microphone to your DV Camcorder, then connect this via firewire into something like Final Cut Pro's built-in Voiceover tool, your camcorder acts just like an audio interface.

I used this technique exclusively whist with a company too tight to provide the correct tools for the job. I'm now independent with an Edirol UA-25 as a pre-amp/compressor. :) It took a couple of goes to sort it out, but works great with Boom Recorder (Mac software recorder that does time-of-day timecode into FCP):

http://www.vosgames.nl/products/BoomRecorder/

Its other great trick for VO work is that a simple keypress splits your current live recording, increments the 'take' label and continues to record to a new file without losing ANY audio - if talent fluffs and starts to retake, hit the key. No 'snip snip snip' later on, just top and tail.

[Loren Miller]

[This may sound cheap and miserley, but if you connect your microphone to your DV Camcorder, then connect this via firewire into something like Final Cut Pro's built-in Voiceover tool, your camcorder acts just like an audio interface.]

I'll go Matt one better. Get an Apple iSight. It's a FireWire camera-- and a darn good 48K mike with - to my ear- a very respectable signal. It's recognized immediately within FCP's VO tool.

Hang it upside down on an IKEA desklamp on its magnetic stand. This gives you hands-off flexible positioning.

And while it doesn't have that cool split-track/continue-recording feature Matt indicates, you just mark an In and Out in the timeline, go Shift-C to begin recording, and ESC to stop. Clip name increments by one and starts on a new track.

I have used this in two ways, 1) Scratch narration for the docs I cut; 2) Low budget FireWire feature ADR. I place the actor right in front of the screen so he can rehearse to his own lip movements or the action. He dons iso cans so he can hear the 5-second run-up but we don't, and he lays down as many tracks as needed.

In a quiet room it works a peach!

- Loren

[Steve House]

Originally Posted by Matt Daviss

...

Its other great trick for VO work is that a simple keypress splits your current live recording, increments the 'take' label and continues to record to a new file without losing ANY audio - if talent fluffs and starts to retake, hit the key. No 'snip snip snip' later on, just top and tail.

Some programs- thinking Nuendo as an example here - support punch-in and punch-out recording with layering of multiple take. To replace a fluffed up part, you mark your in-point and out-points for the replacement. Start playback and at the in-point it switches to record, switching off again at the out-point. The new material overlays the old but doesn't erase it. Need a second try, just repeat the process. You end up with the multiple takes in a stack, the one thats audible in the final mix being the layer on the top of the stack. But you can mix and match if needed, picking this word from that take and these syllable's from that one to assemble the version that will be used in the end.

[Jon Fairhurst]

Vegas and ACID allow the same. They're great for tracking, whether recording a voice-over, or a music track. You can set the punch-in, punch-out points and set a loop with as much pre-roll and post-roll as you want. Hit REC and do as many takes as you want, hands free. When done, hit the spacebar, and choose the best take.

[Tim Gray]

Originally Posted by Jon Fairhurst

Whatever you use, make sure that you have access to a compressor effect.

Here's a simple voiceover trick: put your voiceover on two different tracks. Leave one natural. Put a compressor on the other and compress the snot out of it. (Like 30:1 with a low threshold)

Yeah, I think it's commonly used in music production - parallel compression.

Not a bad trick to have up your sleeve, but in my experience, the no holds barred best compressor is manually riding the fader (virtual or otherwise). With DAWs nowadays, you can really do a good job with your mouse of smoothing out offensive peaks without killing the signal.

[Jon Fairhurst]

Originally Posted by Tim Gray

Yeah, I think it's commonly used in music production - parallel compression.

Not a bad trick to have up your sleeve, but in my experience, the no holds barred best compressor is manually riding the fader (virtual or otherwise). With DAWs nowadays, you can really do a good job with your mouse of smoothing out offensive peaks without killing the signal.

Tim,

I agree completely for on-screen dialog. Just tweak the envelopes. You can take down the odd offending peak and boost the empty syllable.

The parallel compression thing is specifically for BIG VOICE voice-overs. I recently recorded myself doing a Hollywood trailer, super-breathy voice-over and the results really exceeded my expectations. (Real talent would have done it much better, but not within my budget.) If you're going for the Don Pardo sound, don't just ride the faders, use the parallel compression trick. (Don must have a compressor built into his vocal chords!)

Here's a trick for on-screen dialog normalization: Copy your dialog track onto a second track. Apply very conservative noise reduction (only if needed) to the main dialog track. Apply aggressive noise reduction to the second track. Set the envelope of the first track to nominal, and the envelope on the second track to zero. When the dialog is too faint, mix in the second track.

This technique keeps the ambient noise consistent. Gaining up your dialog doesn't add any appreciable noise. Also, you always have some relatively untouched dialog as your base, so it doesn't sound processed or garbled.

I did this on our 48-hour film project entry this past weekend and I was able to get very nice, even, clean results - fast!

[Jarrod Whaley]

That's a pretty good idea, Jon, thanks.