Discussion in 'Recording' started by Doublehelix, Mar 23, 2002.

  1. Doublehelix

    Doublehelix Well-Known Member

    Oct 7, 2001
    I am in process of recording a very talented female vocalist who has a bit of lisp. I know there must be a way to deal with it, but I am clueless. I am assuming that it must be similar to dealing with sibilance, and maybe the use of a De-Esser...but I am also clueless as to how to use a De-Esser (I know it is basically a notch filter). I have the Waves Renaissance De-Esser and I'm pretty sure I have their regular De-Esser. I have tried using it with the standard "female vocal" preset, but it doesn't do anything that I can notice. She does have a pretty low and deep voice. If I understood more about how these things work, I might be able to tweak it a bit...but it is all Greek to me, so I am at a loss as to where to even start with the tweaking.

    Any help would be appreciated.

  2. knightfly

    knightfly Active Member

    Jan 18, 2002
    Disclaimer - I have never played with this either, but have done a bit of reading on the subject. I am also not familiar with the waves stuff. That said, here's how a hardware de-esser works (sort of) any decent compressor that has a side chain input can be used as a de-esser - dedicated de-essers seem to have more goodies, but I'm not sure what they are. (Lotsa help so far, huh?) Anyway, to use a compressor as a de-esser, you feed the side chain output into a parametric EQ, then feed the EQ back into the sidechain input. This allows you to selectively compress the main input as a function of frequency. You then dial in the offending frequency by ear (usually) by adjusting the gain/freq/Q of the equalizer to target just the sibilants of the program material. Once the settings are right, when the sibilants happen, they are boosted by the sidechain EQ more than the rest of the signal, so the sibilants (only) are compressed more than the rest of the frequency band.

    If you want to "Re-ess" instead of "De-Ess", you would need to set up the EQ on the sidechain to CUT the (non-existent) frequencies, so they are compressed LESS than the rest of the frequency bands. (Whatever you do to the side chain is reversed in the output signal.)

    Probably the best way to do this is have the lady record a track and go away (unless she's your wife, then it becomes even easier, just pretend you need her there and you'll not be bothered) Then run the track thru whatever SW or hardware compressor/de-esser you have, and see if it lets you CUT a parametric band on the side chain. If so, do a cut of about 6-8 dB, Q of maybe 3-4, starting at about 2 kHz and sweeping up to maybe 10-12 kHz.. Play the track, grab the freq control and sweep slowly as the lisping happens, and dial things in so what little sibilance that exists gets boosted without the rest of the band being affected. Like I said, this is from (old) memory, not experience, so you'll also have to play with compression settings. The main thing to remember, is that the sidechain does the opposite to the output as you do to the sidechain.

    If none of this works and you're desperate, try a BBE plugin or another "exciter" of some kind.Good luck, and remember: Anything you do right is MY credit, anything you do wrong is Opus's fault... Steve
  3. GentleG

    GentleG Guest

    If none of this works ... and remember: Anything you do right is MY credit, anything you do wrong is Opus's fault... Steve

    I'm sorry Steve I didn't credited you for everything I did, but I sure as %^&* did blame Opus if it went wrong ;-)
  4. knightfly

    knightfly Active Member

    Jan 18, 2002
    Hey, if he didn't think he was so big and badass, he woulda taken the name "couplet", or at the most "verse", "chorus", but hey, "Opus"??!? C'mon, he's got it comin'... Whassa matter, Ope, mommy won't letcha come out and play? :=)

Share This Page