Creating a BIG voice for speech - announcer type

[Dr Gruv]

looking for the parting of the red sea type of thing.....

beside compression and eq and slight delay - what specific settings and other plugs do you use

i have waves plat. and psp delays and the new waves ssl...

could you be very specific as to settings on compressors or other plugs you suggest - thanx

also what would the channel plugs and mastering plugs be?

a long time ago there was an article about a guy in nashville and the wave settings he used were published in an article (that i can't find) in my magazine pile...

anyhow --- thanx

[theblue1]

Proximity, proximity, proximity.

Using a cardioid (unidirectional) mic, the closer the VO artist comes to the mic, the 'bigger' he'll sound. Much of that comes from the proximity effect (the sound becomes bassier as you get closer) but there is also the issue of increased detail (the little things like tongue and lip wetness that create subtle detail and intimacy).

But that's not all of the effect you're looking for...

Once you've got the voice sounding full, you've got to psychoacoustically place it so that it not only sounds full -- but sounds like it's filling up a big space.

To that end you want to very subtly use a little bit of echo and reverb to create the illusion of a big space around the voice -- but one that is filled up with that voice.

So you'll be keeping the voice big, fat, and mostly dry -- but adding a very small amound of fairly long predelay and then just a bit of reverb at the end of the delay. (The long predelay before the reverb creates the illusion in the in the listener's mind that his voice is carrying a long, long way, and then finally bouncing into something and creating reverberation.)

If you use too much delay and reverb it'll just sound like an obvious effect and may well end up making the VO announcer's voice sound smaller or hollow.

So you want to fill up the 'frame' with the VO voice and then just allow a little verb in to define the space around it. If you catch my drift.

Good luck!

[Lagerfeldt]

First of all you really need to start at the beginning. Relying on plug-ins only is a very bad choice if you want quality. Trust me on this.

I think proximity is also a bit dangerous since it can cause some serious low end pops and boomy problems, even with the pop filter. Proximity to the mic can also cause a rise in sibilance.

If this happens try still being close to the mic but slightly turning your head to let air pass just besides the mic instead. This will sometimes help the low end and/or sibilance but still retain a lot of presence in the vocal.

Personally I use cardioid or super cardioid on the mic, and also I activate the low cut at around 80 Hz or below.

Also experiment with the preamp gain/output. Sometimes turning up the gain and lowering the output on the preamp will give more presence as it drives the gain stage differently - but can also lead to unwanted saturation. The opposite - less gain but more output - will usually give a clearer signal. Also try changing impedance or loading on the preamp to see what impact it has on presence.

From the preamp I would recommend going into some kind of compressor or limiter, e.g. an 1176. Using an 1176 - depending on what revision or clone - use 4:1 ratio, relatively fast attack around 15h, very fast release since the 1176 release isn't super fast, adjust input according to amount of limiting.

As a rule of thumb fast attack and medium to fast release is good for vocals. Too fast release and it could start pumping/breathing too much, but set it fast enough and it'll bring up details and endings of words which is good for intelligibility. Fast attack will help control explosives and make the vocal clear yet soft. The higher the ratios, the higher the threshold is another rule of thumb.

From there on you go into your DAW and from here on it could be time for de-essing. However, I recommend trying to avoid de-essing if possible. I would recommend Sonalksis DQ-1 or CQ-1 for de-essing, perhaps Waves Renaissance De-esser set to narrow mode (2nd choice). Good voices will control their sibilance and the rest could be removed or ducked via volume automation, which always sound more natural.

Personally I then go for the equalizer and cut some low end again, maybe up to 80 or even 100 Hz in some cases, using the HPF with a 24dB/Oct filter. I then use a few parametrics to dip any unwanted resonance in the voice, or caused by the mic or room or any combination thereof. These frequency points will differ a bit depending on many factors. Use relatively high Q values when cutting here. I don't add any top right now. I prefer the Sonalksis SV-517 equalizer when I'm working native or Sony Oxford when I'm in ProTools.

Then I use a multiband compressor (Waves Linear Phase, even though it has slight latency), using it as a low level compressor (upward compression). I don't like the Waves C4 so I can't recommend that, even though I use it when I'm forced.. If you're using the Linear Phase instead, start by setting up in bands 2-5 with threshold around -36 to -40dbFS with a range of 2-3 dB and similar make up per band. The lowest band (1) I tend to bypass or use as downward compression, with no makeup. Adjust the thresholds individually afterwards in bands that are very active so they don't over-compress, and also adjust the makeups individually at this point, maybe to add that shimmer in the highest band or some intelligibility in the area around 3-5k.

Now I add a bit of regular compression using either Waves Renaissance compressor or Sonalksis SV-317. Again, fast attacks and relatively fast release. Ratio could be 2:1 or less as there's already quite a bit of compression going on. Try switching off ARC for a more steady sound, experiment turning on/off electro/opto and smooth/warm.

From there I sometimes (depending on the job) use a Waves SuperTap 6 Delay as a kind of chorus. But to avoid sounding metallic I do the following: load the Chorus preset and pull down the levels on the individual delays to the left, around -20 or less. Then activate all the filters to the right and change them to high pass, and set the freq value to maybe 3.5k. Now the metal is gone and all that's left is a nice and very discrete stereo width. Check the output doesn't clip.

As the final insert I use a Waves L1 limiter doing a very tiny bit of limiting. Do not overdo it here or it will sound crap on air. Try switching off the ARC and setting the release manually.

Finally I might either use a bit of delay or reverb. Be sure to cut lo/hi in the delay. If you're using reverb it should be very subtle and make sure to adjust the predelay so it isn't too fast.

-NOW, this may sound like a lot of processing but this is to show a full chain, and maybe only subtle stuff is done at each stage. Multiple compression is certainly a good way of obtaining level without necessarily sounding overcompressed.

Hope it helps, I'm currently producing the sound design and voice-overs for all the new internet radio channels from the Danish Broadcasting Corporation, and normally also do a lot of regular radio spots and campaigns.

[Dr Gruv]

thank you fantastic details and a great answer - PERFECT - that will work great i'll try it saturday! thanx again for the detailed response

[theblue1]

Also experiment with the preamp gain/output. Sometimes turning up the gain and lowering the output on the preamp will give more presence as it drives the gain stage differently - but can also lead to unwanted saturation. The opposite - less gain but more output - will usually give a clearer signal. Also try changing impedance or loading on the preamp to see what impact it has on presence.

Hmmm. That sounds inverse, to me.

Raising the preamp gain sufficiently will tend to saturate the preamp signal.

Lowering the gain will tend to deliver less (or no) preamp saturation but more self-noise. (Obviously, things get more complex if you have a multi-stage pre, or a pre with built in EQ or compression, etc.)

A well designed converter should have an analog input stage with enough headroom to handle an input signal well over what will deliver digital zero without any saturation -- if it doesn't, it is, by definition poorly designed.

Seems to me.




PS -- Good points on the dangers of proximity! (I guess I must have figured it went without saying. But really, in a good explanation, little should go without saying. People can scan past what they already know.)

[hourglass]

Originally Posted by Dr Gruv

looking for the parting of the red sea type of thing.....

Why not just cut right to the chase?

Have Bob Oakman do your voice over for you.

He's the voice of all the Guitar Center ads and the ubiquitous "sunday, Sunday, SUNDAY!!!..." monster truck radio spots.

Hit him here:

contact@bobsbus.com

ryan

[Lagerfeldt]

Originally Posted by theblue1

Hmmm. That sounds inverse, to me.

Raising the preamp gain sufficiently will tend to saturate the preamp signal.

Lowering the gain will tend to deliver less (or no) preamp saturation but more self-noise. (Obviously, things get more complex if you have a multi-stage pre, or a pre with built in EQ or compression, etc.)

A well designed converter should have an analog input stage with enough headroom to handle an input signal well over what will deliver digital zero without any saturation -- if it doesn't, it is, by definition poorly designed.

Gaining the input a lot will tend to saturate the signal. The output will usually not do so when turned up (in the same way at least). This translates to: use the preamp input stage with a reasonable amount of gain and then use the output for output. Important to mention, since some people might instinctly just raise the input stage and then lower the output in order not to clip into the DAW or whatever is the next stage, instead of getting a good input/output balance in the preamp. Some preamps behave somewhat linear when gaining the input while others will have a logarithmic type staturation that suddenly sound real nasty (or good, like great tube equipment).

Interestingly even a complete tube preamp/eq/compressor can sound very different at the input and output stage. For instance the Gyratec X adds fantastic harmonics, low end and top brilliance at the input stage, while the output stage when pushed beyond unity will start saturating in the mids, somewhat closer to solid state distortion.

As for the converters, a well designed converter still can't do anything about a signal past 0 dBFS, it will always clip. Some converters have a soft clip or limiter, but that's something else. The signal will always clip as there are no more digits to describe a full scale signal going beyond its range.

[RadioMoo]

Originally Posted by Dr Gruv

looking for the parting of the red sea type of thing.....

Hire me.

[Vocalvoodoo]

Originally Posted by RadioMoo

Hire me.

Or me!

[Lagerfeldt]

Josh, could you send me a price list?

[Henchman]

No way you're going to get that sound, unless the vocal talent has the voice.
No matter how many plug-ins you pile on top of it.

[joeq]

I agree with Henchman (call the Guinness book!) Some people have that sound in their voice. Get the right announcer. Many of these guys have setups for remotes, so they can be anywhere.

This is a U195 into a Pendulum MDP-1 - no EQ no comp no nothing

[Vocalvoodoo]

Originally Posted by joeq

I agree with Henchman (call the Guinness book!) Some people have that sound in their voice. Get the right announcer. Many of these guys have setups for remotes, so they can be anywhere.

This is a U195 into a Pendulum MDP-1 - no EQ no comp no nothing

That's a hell of a voice that guy has. thumbsup

[joeq]

Originally Posted by Vocalvoodoo

That's a hell of a voice that guy has. thumbsup

It's Ossie Davis. He sounded like that in the room. All I had to do was stick the mic in front of him.

[Mark Vain]

Don LaFontaine


Mark

[Lagerfeldt]

Originally Posted by joeq

I agree with Henchman (call the Guinness book!) Some people have that sound in their voice. Get the right announcer. Many of these guys have setups for remotes, so they can be anywhere.

This is a U195 into a Pendulum MDP-1 - no EQ no comp no nothing

Very deep, but he sounds like he needs to get in better shape with all that puffing. Oh, and clear his throat. ;-)

[joeq]

Originally Posted by Lagerfeldt

Very deep, but he sounds like he needs to get in better shape with all that puffing. Oh, and clear his throat. ;-)

Well he was in pretty good shape for 87

[joeq]

Originally Posted by Mark Vain

Don LaFontaine


Mark

very cool. The video "5 men in a Limo" with the 5 announcers going to an awards show was hilarious.

I saw a comedian on TV who "did" Don LaFontaine for satire of movie trailers, the whole audience "got it" but I doubt if very many knew the name of person they recognized so easily.

[moon_unit]

There's a reason why only one or two guys are doing most of the VO's for film trailers ... and why only a small handful are doing the VO work for most of the car comercials, etc.

It's because they have the voice for it.

What you're asking for is akin to asking: "How do I get my voice to sound like __ insert multimillion dollar recording artist or vocal talent?"

If you talk to one of these VO guys in person or on the phone ... that's what they sound like. They just have these unbelievable resonances, mostly in their chest region. You can train your voice to get somewhat closer to that ... mostly by relaxing your vocal chords and learning how to deepen your voice. But it's not going to happen by using a certain type of mic or EQ or compression technique. Sorry.

[zeneca]

Thanks for these tips. Worked out pretty neatly!

[Lagerfeldt]

4 year old thread you relived here... but I'll answer anyway ;-)

The chain was an example that offers a variety of choices, and as already stated at the end of the thread, some stages could be quite subtle. Always skip any stages you feel is superfluous.

I use the Alchemist today for multi-band BTW and no the Flux Limiter (mostly).