"crunchy" voice EQing (first heard on iphone 4)

so i saw/heard an ad for (forgive me) Goldman Sachs after a TED talk, and it reminded me that i was trying to figure out how to make the vocals sound like it…


i also recall that it stuck out to me when the iphone 4 first came out, that the vocals sounded similar to this as well (i was working at a call center at the time). either way, i’m trying to recreate it.

any ideas? i think it has a bass boost at 150hz or so. also, i think it’s dynamically compressed. there’s also a “crispy, crunchiness” in the … i think sibilance is the right word (kind of in the back-of-the-throat sounds like “Ks”) but after fiddling around with EQ for 30 minutes, i’m stuck.

The commercial features correctly recorded sound. All those people sound like that if you walked up to them on the street.

That’s not to say it’s easy. If I tried hard enough I bet I could find the professional recordist who shot the voices. I didn’t see a microphone, so it was probably done with a fishpole and shotgun with a dead cat.



Yes, it’s been through post production to make them all match and maybe a little level shifting and possible compression, but the object is natural voices and they pretty much hit it. One of the advantages of top quality microphones and capture environment is little or no noise. That means whole equalization and processing worlds open up to you because you’re not constantly trying to fight noises. This is a major problem with home recording.


How are you trying to do it? There is another poster that managed to crank out very clear recordings (attached). Oddly, he thinks they’re damaged and wants to equalize them too.

As I posted to him, acting is a very big part of this. It’s not just compression and Eq. I bet they shot those people about a hundred times to get the clips they eventually used.

“OK, this time can you do it and sound more authoritative? And don’t mumble so much. It’s hard to pick up your voice.”

Still rolling and… Action!


aren’t the “K” and “P” syllables kind of exaggerated in a bit somehow? (as well as dynamics compressed) or is the former sound just the mic?

one thing that seems like how it was processed was that the vocals (in the ad) seemed cut off around the 5-6kHz area, but the “K” and “P” and to a lesser extent “S” syllables allowed to get thru in a muted kind of way.

what i’ve been trying to get to sound the same is pretty professionally recorded audio (Skyrim’s dialogue), tho sadly it’s been lossy compressed, bitrate wise.

any idea what kind of compression? (and is it doable with whatever comes with Audacity?)

IMO yes, the attack on the plosives such as “K” “P” “T” is too pronounced : too clicky, a bit Khoisan.
That can be reduced by using a plugin for Audacity called Paul-L’s De-Clicker

a friend pointed out that a component could be distance from mic coupled with a quieter spoken voice, so i’ll play around with this as well.

in audacity, i think i’ll have to do/work in layers… like a bass boosted layer, and sibilance layer, too (at least until i get super comfortable with custom EQ files)

An improvised pop-filter may be worth a try to tame the plosives like K T P B & D …