If you find that sibilance is a problem on your vocal recordings, a de‑esser such as Drawmer's MX50 can provide a relatively transparent solution.
I have recorded a large number of vocalists with the following mics: AKG C414 B/ULS, Rode NT2, Rode Classic, Shure SM57 and SM58, Peavey PVM 480. However, I am always encountering a big problem with the unnatural amount of sibilance on my recordings, and those that other engineers have done in the same studios. I've never tried a de‑esser, but I get the impression that these are frowned upon. Are there better ways of positioning the mic for less sibilance? I'd love to be able to avoid recording it, rather than having to process it out afterwards. However, if I can't avoid having to de‑ess, I'd be interested to know what's the best way to go about it.
Alan Matthews
Editor Paul White replies: Sibilance can be a problem with some performers, but it isn't usually associated with a particular studio. Some mics with strong presence peaks can emphasise the problem, though singing over or under, rather than into, the mic can help. The old trick of fixing a pencil vertically in front of the mic grille using elastic bands can help too. In really difficult cases, using a dynamic mic may do the trick.
Other causes of sibilance are using too much HF enhancement, overusing high EQ and using bright reverbs. There are now a number of affordable de‑essers, including particularly good ones from SPL and Drawmer that I've tried. Remember that it's sometimes enough to use them only in series with the reverb send if it's only the reverb that's causing the problem.
Assistant Editor Mike Senior adds: If you don't have a dedicated de‑esser, and you don't really want to splash out on one, then you can often get by if you have a compressor with a side‑chain insert point. If you patch an equaliser into the side‑chain and boost the sibilant frequencies, then the compressor will duck the level whenever there is appreciable energy in that frequency‑range.
Better still, if you have access to a multi‑band compressor, this might help you achieve better results than a traditional de‑esser. If the frequency crossover points of the compression bands can be adjusted, then you can often acheive de‑essing by focusing one of the bands on the most sibilant region — the other bands can be bypassed. The advantage of this is that whenever the compressor is active, it will only dip the offending bit of the sound, without compromising anything else.
There are a number of other things you can do to reduce the sibilance of vocals after you have recorded them, in addition to using a traditional de‑esser or a multi‑band compressor on the main vocal channel or its reverb send. If you are using any sort of hard‑disk based recording system, you are often provided with fader automation which can be used to manually implement much more subtle de‑essing than would traditionally be possible. Just go through your final vocal, ducking the fader level a few dB at every 's' — this might be time‑consuming, but at least you'll know for sure that it'll only duck the sibilance, no matter how hard you decide to do de‑ess. Another interesting idea, posted on the SOS web forum by Andre Bergeron, is to pitch‑shift the harsh start of any over‑sibilant sounds (using an off‑line process) down in pitch in increments of a semitone until they become smoother.