While I agree mainly with Mr. Jim F., (and some others as well),
I would like to mention some techniques, that were not discussed in this topic
yet.
As someone (was it Bob Ludwig probably? 🙂 ) said: the pencil tool is still a tool to use
Well, he probably said it with his own words....
Soooooooo, here is what I call "manual de-essing", it probably needs some kind of experience (as a result of certain ammount of practice):
- first copy the vocal region(s) to another track, save the copied files as new audio files and mute the original files
- I open the vocal region in sample Editor and make Catch clock position and zoom to the level which allows me to see the problematic ssshhh and pppphhh (if SSS and SSSHHs are your only problem, you are ****** LUCKY 🙂 sssshhhhhhht
-hit playback and as soon as my ears get hurt (almost) by ugly frequency attack, I can also see what it was. So I mark that piece of sssssssshhhhhhhhhhhit and change a gain - usually 2-3 db (remember this is done in Sample Editor). It depends on ammount of compression that is applied to the vocal track of that marked piece.
-sometimes I choose "Silence" for a short pieces which are (unwanted) artefacts made by singer's tongue, lip or whatever, that can appear very shortly AFTER or BEFORE the word and which may produce a very huge CLICK if fair ammount compression is applied.
-I keep "cleaning" the track untill I am done 🙂
Now - you may ask, if I do it with effects like EQ and compressors inserted to the channel.
I answer - yes. And sometimes no. I may say that sometimes when tracks do arrive, I recognize that there are too much of those ssssshs and phhhhps etc. and it is sometimes one of first things I do - even when no plugin is inserted yet...
While I agree with what Jim said about stepping a vocal talent back (turn his head a bit), choosing a different Mic and yadadada, I am (and I think I am not alone) often in situation, when I do not have another (better) Mic to choose (or better Vocalist to record 🙂 )
So I learned (god dmned - no - I am still trying to learn) to use de-esser during recording. I mean hardware deesser. slightly ofcourse. I am also found out that having a finger on a fader/knob during recording can help alot. Like sometimes you can say a million times, that if that phrase is about to come, he (singer) should better turn his head a bit or step back etc. - but that is probably not the right thing to do during a REC session... As if the talent does not know this yet, he won't learn it during a sesion and his preformance will be worse - much worse. So I rather have a finger on a fader and / or on a knob of a hardware compressor/de esser and when those sssssshhhhhhhhhiiiittttttttthhhhhhyyyyyyy passsages do arive, I try to react (with certain ammount of lookahead 🙂 🙂 🙂 ) with my fingers to compensate - just to not let deesser eat all of nice "airy" sound...
And I also use more de-essers on one (problematic) track. Two, three - no problem.
If I have to compress the vocal track heeeeeel of A LOT, than first plugin in a chain is usually
UAD deesser. After that I use comp/EQ (mainly SSL Duende Channel - so the order depends on situation). I usually insert another deesser after comp/EQ! And put another compressor ( uh - is 1176 a "compressor"? - or La2a...) after that deesser.
Another old trick is to use a parallel compression. In this case I "clean" the vocal track that is about to be heavily compressed from esses almost to zero. Because for this track I use a "dolby effect" on focusrite VoiceMaster Pro. It is a really nice effect but it produces heavy artefacts from shhhhhs and pppppphhhhhhs (depending on setting). But with almost no esses and other problematic "ppphhs" I can apply this compression fx really heavily and blend this track with the original track after...
finally the best deesser (in one situation 😉 ) was (after I spent hours of "pencil" clening and usng software deessers and trying to ride volume - but nothing satisfied me ) this: I used the Focusrite RED3 (hardware) compressor as a frequency dependant compressor (it has the key input for the "frequency"). It cleaned the tsound easily and nicely, much better than any other (software) plugin I tried. But to be honest, I am not doing it very often, as UAD-1 deesser serves me well. And I am quite lazy sending a track out of DAW and record it back and allign it to it's original position - too lazy for that to do it often.
I know some guys who use Sonalksis stuff (their MultiBand dynamics plugins like dynamic EQ or comp) for deessing and they are satisfied with those "multiband" deessing techniques.
Finally I would like to mention, that plugin automation is a very big advantage that we have. As Those ugly frequencies do not have to stay be the the same during the song. Also the ammount of deessing does not have to stay the same during the whole song. Plugin automation possibilities can be the answer for the lack of the "airy" sound that can be "eaten" by a deesser.
Message was edited by: Diamond Dog
