Make Mixes That Accurately Translate What You Hear In Your Head.
Make Mixes That Accurately Translate What You Hear In Your Head.
203K+ subscribers
When trying to make your vocals sound clean, use a high-pass filter, try multi-band gating to increase the dynamic range while lowering noise, and ensure that your temporal effects are in-time with the BPM of the track. Additionally, try to control your sibilance with clip gain.
The easiest way to clean up your vocal track is to attenuate your low-frequencies with a high-pass filter, up to around 150Hz, but lower can work well too. This will quickly attenuate a good amount of background noise, mic rumble, unwanted plosives, and more.
I’d recommend using a low-latency linear phase setting to avoid phase changes to the low end.
Although unorthodox, you can use a multi-band downward expander to attenuate the signal whenever the vocal isn’t loud enough, in turn reducing background noise and increase the vocal’s dynamic range. A few dB of attenuation between passages can make your vocal sound cleaner and more distinct.
Use a multiband compressor or dynamics plugin for this job.
Out-of-time reverb can make your vocal sound cacophonous as the reflections will clash with the in-time-instrumentation and original vocal. Avoid this by making your reverb in time with the BPM of your track; 60000 divided by the BPM results in a quarter note in milliseconds.
Use this number or a multiple of it as your reverb’s RT60.
Like out-of-time reverb, out-of-time delay can make your vocal indiscernible and create discord in your mix - what’s worse is that out-of-time delay might cause unrelated notes to overlap. Again, use 60000 divided by your BPM to find a quarter note in milliseconds.
For shorter delays, just divide this number by 2 to get an 8th note, 4 to get a 16th note, and so on.
When compressing a vocal, if you set your release time to quick, you’ll cause distortion to your low frequencies - to avoid this you need to make your release time longer than 1 full cycle of the recording’s lowest frequency. For example, 50Hz takes 20ms to complete 1 full cycle.
Odds are you won’t encounter a singer that can sing lower than this, and with a tenor or soprano, you could use a quicker release time.
Low-level compression is great for adding detail while control dynamics, but it can also raise the noise floor of a recording, making for a less clean vocal. To avoid this use a broadband gate on your vocal to reduce noise and then use upward compression.
Alternative or in combination with a gate, you can cut our sections of blank space in your vocal.
Low-frequencies in reverb can come in handy when adding reverb to bass or kick; however, they can quickly make your vocal sound muddy or boomy if they’re not dialed-in correctly. If the reverb has a control for it, reduce the low-frequency level to what sounds right.
Alternatively, if the reverb was created with a send, you can introduce an EQ with a high-pass filter.
Sibilance is making a vocal sound harsh and shouldn’t be overbearing when trying to achieve a clean vocal; unfortunately, many de-essers have a distinct sound to their compression, which might be unwanted. For a cleaner sounding alternative, find each Ess and use clip gain to reduce its amplitude.
If this is too time-consuming, the Weiss DS1 MK3 makes for an incredibly transparent de-esser.
One good starting point for a vocal EQ curve is to use a high-pass filter up to 80Hz, attenuate 200Hz slightly, boost 500Hz slightly to amplify vowels, then attenuate 700Hz to remove nasal tones. Next, boost 2.5kHz to 3.5kHz for detail, and then add a little air.
Although these numbers aren’t exact for every performance, they work as a good preset for making a vocal sound clean and present.