A small, single‑driver speaker like this Avantone MixCube can be invaluable in checking for inconsistencies in a vocal.
How do I make my ears more sensitive to inconsistencies in the vocal frequency spectrum? In theory, I seem to understand, but when comparing my mix with a reference track, I always feel that my vocal tone needs to sound clearer and more solid. Yet, I’m not sure what frequency range I need to work on.
Anon, via email
Mixing expert Mike Senior replies: A lot of what makes a vocal sound clear and solid in a mix isn’t actually EQ: it’s arrangement, dynamics, fader automation, effects, and how the backing track parts are processed. EQ is only ever part of the solution. But there are a number of tricks I use to assist me in identifying inconsistencies in the vocal spectrum.
Turning the monitor volume up loud is great for highlighting whether certain consonants (like ‘s’, ‘ch’, ‘sh’, ‘f’, ‘t’, ‘k’) sound more or less harsh than others. Once I’ve found one that’s too harsh, I often loop that consonant while looking at a high‑resolution spectrum analyser and checking for spectral peaks that might be the source of the harshness. I’ll then try cutting one or two frequencies with region‑specific EQ to see if that helps. The same tactics work very well for spot‑fixing sporadically harsh‑sounding held notes — usually certain vowel sounds when sung loudly in the higher registers. Again, the high‑midrange resonances that cause this are usually pretty easy to see on a spectrum analyser, once you’ve isolated the offending note.
One of the most common problems is where isolated falsetto notes sound well‑balanced on the small speaker, but far too full/big once listening on your main monitors
Comparing the vocal mix balance on full‑range speakers and a single‑driver midrange speaker (I use my Avantone Mix Cube for this) can highlight where the upper or lower frequency extremes of a vocal are inconsistent. The trick is to get the vocal balance sounding consistent on the single‑driver speaker first, then, if some notes feel too full/thin (or I think of it more as ‘too big/small’) on the full‑range system, you can use EQ to rebalance the vocal’s lower frequencies against its midrange. If some notes feel too bright/dull, or forward/recessed, the ‘air’ frequencies above 8kHz probably aren’t consistent enough. The same comparison between speaker systems can be useful if trying to balance different vocal deliveries (eg. falsetto versus head voice, versus chest voice, versus belting...). If you get the different vocal deliveries sounding even on the small speaker but they then sound unstable on the big speakers, this implies that something about the vocal’s frequency extremes isn’t consistent with its midrange. One of the most common problems is where isolated falsetto notes sound well‑balanced on the small speaker, but far too full/big once listening on your main monitors, because falsetto deliveries usually have significantly less true midrange energy.
Finally, high‑pass filtering a vocal is particularly useful for pop/EDM, where you’re trying to get that kind of consistently dense breathy intimacy. It focuses your ear on the level relationship between the consonants and vowels in that spectral range — the former are often much louder than the latter in project‑studio recordings, but you’ll also find differences in the HF energy and density of different vowels. I’ll usually filter around 5kHz with a fairly steep slope (maybe 24dB/octave), but it depends on the voice.