How To Fix Harsh Vocals In The Mix
Mixing vocals can be a piece of cake when the artist and recording engineer played their parts correctly. But sometimes it can be a head-scratching nightmare if you are dealing with tracks that pierce the ears. Harsh vocals are never good for the end-listener– they aren’t pleasant (borderline painful really) and usually translate or appear worse on certain playback devices.
While harsh vocals are not ideal – in mixing, sometimes this is the reality you are given and when dealing with a tight budget and time constraints, you are forced to work with what you got. In this article, I hope to show you some techniques you can use to correct harsh vocals and get them to sit better in your mix when re-recording is not an option.
Understanding The Problem
The true problem with harsh vocals is that they are caused by specific frequency resonances within the high frequencies of the vocal track. These resonances often don’t happen all the time but rather only when certain consonants, phrases and notes occur within the vocal performance.
Therefore, when approaching harsh vocals, we must be aware of how we can correct this problem without affecting the parts that aren’t problematic. Some, none or all of these strategies may apply – use your best judgement to rationalize which approach to use, and how to organize your session in order to achieve the best results.
Use a De-Esser to Tame Harsh Frequencies
A common quick fix for this issue – or at least a good starting point – is to use a de-esser to tame harsh frequencies on your vocal. This can be applied to individual vocal tracks or directly on a vocal master that controls a group of vocals.
Waves Renaissance DeEsser – a popular de-essing plugin
A de-esser is essentially a form of multi-band compression that focuses solely on compressing the frequency range that typically contains “S”, “CH” and “T” sounds, often between 2-10khz but potentially higher or lower depending on how you set it up.
We can use a de-esser to reduce the volume of these harsh moments only when they occur. This would leave the vocal primarily unaffected when there aren’t any harsh moments, and only have it engage and reduce when we really need it to.
You may also use several de-essers in series to tame the harsh material more moderately and in stages, as opposed to using one de-esser aggressively to treat the issue. By distributing the de-essing work amongst two or more de-essers in series, you’ll split the workload and can tame the harshness in your vocals more transparently.
Often – I will de-ess directly on an individual vocal track and also on the overall vocal master that affects all vocal tracks if I see fit. This helps as sometimes I may tame the harshness on one individual vocal track, but when combined with backgrounds and the other tracks the harsh frequencies build-up again at the master level and continue causing problems. De-essing the vocal master will help correct this issue if and when it presents itself.
Another helpful read for more information on de-essing: “What Is A De-Esser and How Does It Work?”
Use a Narrow EQ to Cut Harsh Frequencies
EQ can be a great tool for reducing specific harsh frequencies overall. I liken EQ to having a volume fader on every frequency that makes up a sound. Therefore, it can be used to reduce unwanted frequencies and increase desired ones as well.
Using a narrow Q, sweep across and find harsh resonances in the high-end (usually between 2-10khz) and reduce them as you see fit using your ears’ judgement. Keep in mind that doing this will affect that frequency’s presence throughout the entire song. For example, I may notch out 3db of 4.5khz and find that it corrects my problems during harsh moments in the verse, however, when a key change occurs during a chorus, losing 4.5 khz could negatively affect the overall presence of the vocal during those moments. Because of negative side-effects like this, I resort to EQ after I’ve explored my other corrective options.
It may be also good idea to use automation to bypass the EQ when it is not needed, that way you effectively reduce the harsh frequencies only when you need to.
Use a Focused Dynamic EQ
Dynamic EQ is my second favorite method for taming harsh vocals. As I previously mentioned, using regular EQ can negatively impact the source material as it affects the sound consistently throughout a song, even at times when you may need the frequencies you are cutting. Automating the EQ can remedy this problem but that is also more work to set up properly. I propose another solution: dynamic EQ.
Brainworx Dynamic EQ – a popular dynamic EQ plugin
Dynamic EQ works the same as traditional EQ with one added twist: it only works when the frequency you are trying to adjust exceeds a pre-determined volume threshold. This is great, as it will only attenuate a frequency when it becomes too loud in the track as determined by you.
I will often set a dynamic EQ on individual vocals and have it reduce harsh resonant frequencies only when they become too overpowering on specific phrases or words. This leaves the other moments of the track unaffected and doesn’t cause problems when key changes occur.
Use Clip Gain and Volume Automation to Manually Reduce Volume
In my experience, I use this approach the most when dealing with harsh frequencies as it often yields the very transparent results when executed correctly.
If you’re using a DAW like Pro Tools which allows you to adjust an audio track’s clip gain – you may crop out harsh moments on a clip, and manually reduce the volume using clip gain to reduce harshness yet maintain the track’s presence in the mix.
When doing this, keep in mind that clip gain adjustments will occur at the pre-FX level and will affect your track’s input into existing processing such as compression, leading to your compressor (and other FX) now reacting differently to the source material.
The other approach involves incorporating volume automation. Using this strategy won’t affect your compression or any processing on the channel as it occurs post FX, and instead allows you to control the overall volume of that track after processing. This can be helpful when using multiple tools to tame harshness such as EQ, compression and de-essing in order to help the vocal cut through the mix.
Use a Multi-Band Compressor to Squash Harsh Frequencies
If you do not have a de-esser, or if you aren’t able to get the results you want from the de-esser, you may use a multi-band compressor to achieve similar results.
FabFilter Pro MB – a popular multiband compression plugin
Enable a multi-band compressor on your source material and create a frequency band within it that focuses on the harsh frequencies. Once you’ve got your band focused on the harsh vocal frequency content, set your threshold to attenuate the band whenever the volume of the harsh parts exceeds your threshold. This will allow the frequencies to travel through and only be reduced in volume when they become too overpowering.
Use a Compressor with a High-Pass Filter Enabled
If you don’t have access to a de-esser or a multi-band compressor, (or just want to switch things up) you can still make use of a regular compressor with a high-pass filter. Since harsh frequencies are often found within the high-end of a track, you can set up the high-pass filter of the compressor to a specific frequency just below where the harsh frequencies reside. From there, the compressor will focus solely on compressing all frequencies above that filter point, which should include the track’s harshness.
This will yield similar results to a de-esser and multi-band compressor, albeit with a bit less control as you will be compressing and affecting all frequencies above the high-pass filter setting, as opposed to a specific frequency range as determined by you.
Tweak The Chain, Use Good Vocal Technique & Record Them Properly
This could really be step one. As much as we look to mixing as a place to correct problems – the real culprit of harsh vocals tends to be a few factors. For your artist, things like their natural tonality; vocal delivery technique and recording position on the microphone will affect sibilance. For the engineer, the vocal recording chain may be causing a build up of high frequencies leading to additional sibilance issues too.
While it’s challenging to be aware of each of these and correct them before recording, it’s essential to do so in order to reduce the amount of work you will have to do afterwards while mixing, and lead to a cleaner song overall.
You may find using one or two of these strategies will help you fix harsh vocals in your mix. In my case – some deliberate clip gain and volume automation paired with de-essing and dynamic EQ often solves my harsh vocal issues when they come up.
Understanding each of these processes and knowing when to apply them properly in your mix will make all the difference in solving your harsh vocal issues.