Superior Voice-Overs

[Dana Salsbury]

I'm filming a movie trailer/ad to be played at a local theater, and want to do some voiceovers. I have Zoom H2s, a Rode shotgun, Audio Technica lavs, but not the expensive studio mics. I'm guessing the Zoom is what I should use. I have the options of 96kHz/24bit wav, 320kbps mp3 or vbr mp3. (Of course I have lesser values of each). I'm thinking that for the best sound I should use a closet or hang a comforter to eliminate echo. It's a very short voice-over section, but should I rent something, or do I have the bare essentials to do it right? I'm also guessing the standard voice-over effects are something I could pull off in FCS2?

[Marco Wagner]

A closet is a great place. I've done it in the closet and also in the open. I created a "pop" guard from my wife's stocking and some flexible metal. That is one thing to consider is the "pop" from the voice (and its air) to the mic.

[Jarrod Whaley]

As always, what you do with your gear is more important than the gear itself. I may be summarily beaten about the neck and face for what I'm about to say (wink), but yes, I do think you can pull off a very nice-sounding voiceover with the equipment you already have.

However, you can probably produce an even better voiceover with a large-diaphragm mic, some of which are not really that pricey. The AT2020 (which is really sort of a medium-sized-diaphragm, but it actually sounds surprisingly good for its cost) at about $100 springs to mind, and as you suggest, you can probably rent something even better for even less than that.

Also bear in mind that the most important part of getting a good voiceover recording are the vocal characteristics, experience, inflection, etc. of the person doing the speaking. If you get that part of it down, you're 75% of the way there.

There is lots of information in this forum regarding voiceover recording. Do some searches and lots of reading, and you'll learn a lot of really useful things. There are some very, very knowledgeable and skilled people who read and post to this forum regularly (hello, Ty, DSE, Steve, et al.). Listen to what they have to say, and you won't be disappointed.

[Bill Davis]

Originally Posted by Dana Salsbury

I'm filming a movie trailer/ad to be played at a local theater, and want to do some voiceovers. I have Zoom H2s, a Rode shotgun, Audio Technica lavs, but not the expensive studio mics. I'm guessing the Zoom is what I should use. I have the options of 96kHz/24bit wav, 320kbps mp3 or vbr mp3. (Of course I have lesser values of each). I'm thinking that for the best sound I should use a closet or hang a comforter to eliminate echo. It's a very short voice-over section, but should I rent something, or do I have the bare essentials to do it right? I'm also guessing the standard voice-over effects are something I could pull off in FCS2?

Dana,

Here's the central issue. What's important to you?

If it's the RECORDING - then concentrate on the recording gear.
If it's the Performance - then concentrate on the talent you use.
If it's the effectiveness of the production to generate results - then concentrate on the CONTENT of the spot.

You've got to be careful because there's a whole bunch of truth in the saying that if the tool you have is a hammer - your problems start to look like a nails.

Microphones don't make outstanding recordings. Only HUMAN BEINGS can do that.

As the producer, YOU have to determine and secure talent sufficient to produce the results you want.

Just realize that there's not a microphone that's ever been invented that comes with talent built in.

Plus, after more than 2000 professional VO gigs, I can assure you that there aren't any "standard" voiceover effects. Sometimes some compression. Sometimes some other processing. But the formula you hear on 99% of the narration VO work you hear out there is to simply get a good recording of a good voice doing a good delivery of the content and then DON'T MESS WITH IT.

For what it's worth.

[Dana Salsbury]

My gosh.

Bill, I'm surprised about the effects thing. It seems like there is quite a bit on movie voice overs, especially reverb. How about EQ settings? The lows always seem so punchy. Then again, it has a lot to do with the resonance in the voice -- back to the talent. Since I'm the talent, I'm going to do some more reading.

When I sang on an album we layered two cuts of me singing the same thing. Do voice-over pros do this? Should I do it if I don't get the resonance that I'm after?

[Bill Davis]

Originally Posted by Dana Salsbury

My gosh.

Bill, I'm surprised about the effects thing. It seems like there is quite a bit on movie voice overs, especially reverb. How about EQ settings? The lows always seem so punchy. Then again, it has a lot to do with the resonance in the voice -- back to the talent. Since I'm the talent, I'm going to do some more reading.

When I sang on an album we layered two cuts of me singing the same thing. Do voice-over pros do this? Should I do it if I don't get the resonance that I'm after?

Dana,

You're right that the "sound" of movie VO work is typically pretty heavily compressed and "IN YOUR FACE."

But those are professionally recorded VOs. They're caught in totally soundproof VO booths by top quality gear. The signal to noise ratio is STELLAR to begin with. So you can compress the bejezus out of them without risk of also compressing crap like cable noise, hiss, room AC, and the thousand other shocks that typical home studio recordings are heir to.

It didn't sound to me like you're working at that level of sophistication. So you're gonna take a track with a moderately decent s/n and if you compress the heck out of it, you're gonna end up totally MAXIMIZING everything that's LESS THAN perfect about it.

And that extra reverb that might give a GREAT VO recording a bit of extra depth - will just further MUDDY up an average one. Particularly if the talent doesn't bring pro quality articulation to the job.

I'm not saying you can't do this well. And I'm certainly not saying you shouldn't try it your way.

I am saying that the REASON that major movie trailers sound the way they do isn't because they perform some magic tricks on mediocre raw materials to make them suddenly sound PRO.

They sound PRO because they ARE pro at every step of the way. GREAT voice, GREAT read, GREAT recording, Great spot specific tailored processing.

I've never seen a "VO" button on an effects processor. Have you? Why not?

Here's my answer.

If "GREAT VO SOUND" was primarily a function of electronics, someone would have long ago put a button on a $500 box that said PUSH FOR MOVIE TRAILOR SOUNDING VO - and gotten rich off it.

On your final point, no, I can't remember in the last 30 years anyone "doubling" a voiceover track I've recorded. That's not to say no-one has, because for the first 20 years I worked, I was just the talent and the recording engineers did the final mixes. But with radio station compressors and limiters trying to pump up and pump out absolutely the most competatively "loud" sound possible - the goal was always to start out with a clean, clear, perfectly articulated track as the base. So that when the electronics made it BIGGER it was the SIGNAL being enhanced, not the noise, or the reverb, or the compressor breathing, or anything else.

YMMV.


BTW, since you're in the Valley, if you're just doing this for your own project and it's not a big "for profit" thing, you're welcome to come over and borrow my VO booth if you want to try to create your own version of the big voice thing.

Take care.

[Chris Swanberg]

Gee Bill, what a generous and kind offer you just made to Dana. This site has the greatest folks!

[Dana Salsbury]

I'll PM you on the booth. That's very kind.

Compression is huge, I concur. FCP has a compressor/limiter. What would be the difference in a software compressor/limiter vs a hardware solution?

This is more of a general sound question, but when I have a low sound level from a mic for whatever reason I double or triple it up in FCP, which both increases the volume I have to work with, but it also provides a fuller sound. Nobody taught me that, so it may very well be a poor solution. What, in effect, am I doing to the sound by layering it that way?

[Jon Fairhurst]

The title of the thread is "Superior Voice-Overs", and for that, you want it all. Lifting words from Bill's earlier post and putting them in the order of the chain: CONTENT, PERFORMANCE, RECORDING, and PROCESSING.

One excellent and easy processing trick is to copy the raw track, apply MASSIVE compression to it and none to the original. Start with 20:1 compression and a threshold of -25db. Fine tune the attack and release times to minimize breathing. Now mix the dry and processed tracks to taste.

This isn't really doubling, since you're not using a separate recording, nor are you delaying one of the tracks. Keep 'em perfectly time aligned to avoid comb filtering.

Why does this work? The compressed track keeps the soft parts from getting too soft. Trailing words, dropped vowels and soft consonants are boosted right up. And the dry track keeps the sound dynamic and lively. Listen to the compressed sound alone and it can be flat and dull. The dry track spices it right up.

You can then bus the two tracks together and EQ, add a snip of reverb, whatever. Control the mix with the bus fader, rather than trying to manage the two parallel tracks. Leave any further compression and limited for final mastering.

Sure, there's a lot of esoteric gear out there, you can use multiband compression and other tricks. But the above approach can work in most any DAW or NLE as a meat a potatoes approach.

Give it a try!

[Steve House]

Originally Posted by Dana Salsbury

I'm filming a movie trailer/ad to be played at a local theater, and want to do some voiceovers. I have Zoom H2s, a Rode shotgun, Audio Technica lavs, but not the expensive studio mics. I'm guessing the Zoom is what I should use. I have the options of 96kHz/24bit wav, 320kbps mp3 or vbr mp3. (Of course I have lesser values of each). I'm thinking that for the best sound I should use a closet or hang a comforter to eliminate echo. It's a very short voice-over section, but should I rent something, or do I have the bare essentials to do it right? I'm also guessing the standard voice-over effects are something I could pull off in FCS2?

Definitely DO NOT use any of the mp3 recording formats. Compression artifacts that would be nearly unnoticable on a typical television sound system or even a home theatre sound system could jump out like a sore thumb when played out high levels in a movie theatre. 96kHz isn't such an advantage over 48kHz but it won't hurt - OTOH, going with 24 bit uncompressed wav is a definite plus.

Consider recording directly into your computer if you have a decent audio interface (as you should anyway so you can properly mix your soundtracks). Jarod mentioned one possibility for a mic to consider - another worth looking into is the Rode NT1a - good sound at a very good price. Add one of the small Mackie mixers and you'll be in business.

[Dana Salsbury]

You guys are excellent.

The Zoom records to an SD card, so there are never any cord connections. If I went straight to the computer I think I'd pick up some noise.

>Control the mix with the bus fader, rather than trying to manage the two parallel tracks.

Would you elaborate. Wouldn't I be doing all three?

[Jon Fairhurst]

Originally Posted by Dana Salsbury

>Control the mix with the bus fader, rather than trying to manage the two parallel tracks.

Would you elaborate. Wouldn't I be doing all three?

I set the mix between the uncompressed and compressed track once with the two track faders and then I don't touch them. I ride the levels with the downstream bus fader when mixing/automating over time with the foley/soundtrack/sfx/etc.

Here's another trick. Voices typically include these critical bands (These are rough guidelines. It varies by voice):
* 250 Hz (or there abouts) - voice fundamental
* 600 Hz - not very important
* 1.2 kHz - consonants
* 2.4 kHz - character
* 5k - 15k - air and sparkle

Setting the level of the fundamental is critical to getting a natural balance. If the voice is hard to understand, boost the 1.2 kHz region. If there are multiple speakers that sound similar, you can enhance differences in the "character" region - or you can smooth a nasally voice by cutting those frequencies. You can add air and sparkle, but if you overdo it you get noise and sibilance.

When mixing with music, cut the music a bit in the voice fundamental and consonant areas. You can boost the music below the fundamental to make up for it. Cut the music in the consonant area to aid understandability. You can add sparkle to both the voice and music as needed.

The character area is a judgment call. You need these frequencies to hear the difference between flute, oboe and clarinet. Whether you give this space to music or voice or both is a judgment call.

As a general rule, use cuts, rather than boosts. Cuts can be narrow and deep. Boosts, if used, should be wide, smooth and moderate.

You can actually be quite brutal with EQ and get a reasonable result. If you hear phasing, hollowness, or resonance, you've gone too far. Whether or not you want it to sound "natural" or "hyper-real", though, is a judgment call.

[Allen Plowman]

I tried recording to PC, and I could not prevent recording my cooling fans. (I do not own a laptop)

[Jon Fairhurst]

Originally Posted by Allen Plowman

I tried recording to PC, and I could not prevent recording my cooling fans. (I do not own a laptop)

The solution? longer cables. Either user a long mic cable and record remotely, or long keyboard/video/mouse cables and move your PC out of the room.

You can also put your PC in a cabinet and/or use near-silent components: http://silentpcreview.com

[Dana Salsbury]

Okay, now I'm having SERIOUS fun! I love adjusting the EQ, but it has always seemed hit or miss, and questionably a time waster in the larger scheme of things. Then I'll come across a frequency that makes it all worth while. This info will tell me what to target. Thanks!