How to give your voiceovers that professional presence in Adobe Audition
Feb 7, 2018
John Aldred
John Aldred is a photographer with over 25 years of experience in the portrait and commercial worlds. He is based in Scotland and has been an early adopter – and occasional beta tester – of almost every digital imaging technology in that time. As well as his creative visual work, John uses 3D printing, electronics and programming to create his own photography and filmmaking tools and consults for a number of brands across the industry.
Share:
No matter how good the equipment we use to record audio, there’s always some room for improvement. Of course, we have to record it properly, eliminate background noise, echo, ground hum and pre-amp hiss, but there’s more to it than that. High-end voiceovers just have a feel about them that isn’t just a plain old voice recording. They have a warmth and richness to them.
Getting that feel isn’t so difficult. Your mileage will vary depending on the source material (your natural voice), but it’s just a few simple steps in Adobe Audition. In this video, Nathaniel Dodson from Tutvid walks us through the whole process.
Nathaniel uses the Rode Procaster microphone for recording his tutorials and voiceovers. It’s a fantastic microphone, essentially it’s an XLR version of the Rode Podcaster. As it’s XLR, it requires some kind of interface between it and your computer. So you’ll need a mixer or some other device that can supply the microphone with phantom power as well as relay that signal into your computer.
The Rode Podcaster, on the other hand, is a USB microphone. So it’s a lot easier to deal with, if recording to computer, but also offers excellent sound. So, assuming you’ve got a decent quality clean track to begin with, this is Nathaniel’s workflow.
Light Noise Reduction
This process isn’t always essential, but it can help to take care of that last bit of noise. Even the best recorders can still sometimes have a little hiss. Or, perhaps there’s a barely perceptible distant hum of the neighbour’s refrigerator. This will take care of it. The easiest way to deal with this is to account for it when you’re recording.
Leave a few seconds either at the beginning or end of your recording. Then, in Audition, select one of those blank areas, and choose Noise Reduction -> Capture Noise print from the Effects menu. Once captured, select the entire audio file, and from the Effects menu, select Noise Reduction -> Noise Reduction (process). A dialogue box will pop up showing various options.
The exact settings will vary depending on the source footage you’re dealing with, but Nathaniel talks about the various options and what he prefers to do.
Normalize
From the Effects menu, choose Amplitude and Compression -> Normalize (process). Again, a box pops up. Check the box that sayts “Normalize To:” and enter -6.00dB. Then hit apply. This will then raise up the volume of the whole track until the loudest part hits exactly -6dB.
Noise Reduction #2
This is basically just repeating the first step. Sometimes, normalising can raise the overall volume of the recording so loud that you start to hear hiss and hum that you didn’t originally notice. So, this allows you to just go ahead and clean that up.
Compression & EQ
This is where we start to actually flesh out the tone a little. So, from the effects menu, choose Amplitude and Compression -> Single-band Compressor. This helps to just even things out just a little more to help lower some of the peaks to create a more consistent overall volume.
Then the EQ. This is where we enhance certain frequencies to give some richness and depth to the voice. So, again, from the Effects menu, choose Filter and EQ -> Parametric Equalizer. Nathaniel starts with the Loudness Maximizer preset.
He then tweaks some things around to get just the right sound that he’s after. It helps to have the audio playing in the background so that you can hear the changes in realtime as you make them. Exactly how you adjust the EQ is going to depend a lot on your voice, the quality of the recording and how you want it to ultimately sound.
Normalize, again
With all of the changes made, it’s time for a final normalize to bring everything back to a standard level. This time, though, instead of normalising to -6dB, we go to -3dB. This will be the final loudness of the audio for the finished production.
Obviously, not every voice or every recording is going to be processed exactly the same way. There might be loudness issues, or you just find that the settings you used on your own voice last time don’t work for somebody else’s voice next time. So, Nathaniel also goes into some of the potential issues that can crop up, as well as how to get around them.
As a bonus tip, he also shows you how you can record the whole thing into a macro so that if you’re doing the same thing over and over again, you don’t need to remember all the steps. You can just hit a keyboard shortcut and boom, processed.
John Aldred
John Aldred is a photographer with over 25 years of experience in the portrait and commercial worlds. He is based in Scotland and has been an early adopter – and occasional beta tester – of almost every digital imaging technology in that time. As well as his creative visual work, John uses 3D printing, electronics and programming to create his own photography and filmmaking tools and consults for a number of brands across the industry.
Join the Discussion
DIYP Comment Policy
Be nice, be on-topic, no personal information or flames.
8 responses to “How to give your voiceovers that professional presence in Adobe Audition”
Pop filters are a wonderful invention too…
Luke Wilson
No need for phantom power on a dynamic microphone like the Rode Procaster. Condensor mics need phantom power.
The only reason to turn on phantom power with a dynamic mic is power a pre-amp like a Cloudlifter or FetHead. These pre-amps sit between the mic and the audio interface. The pre-amp will use the phantom power to boost gain and not pass on the 48V to the mic.
Ashling Smith
I hate my voice. Will this fix that? :)
Probably not the best advice in this one.
What are we supposed to do with this comment? If you’ve got something better, share it. Otherwise keep quiet.
Ok Arthur, you asked for it my friend.
Recording:
1) Spend some time sorting your recording environment. Find a quiet area with as little room ambience as possible. Some moving blankets and the likes will work wonders. We regularly set up moving blankets on stands when we record dialogue on location for clients.
2) Choose the right mic for the voice. My voice has a lot a lower mid-range, so dynamic broadcast mics aren’t very flattering in my situation. For example, we recently purchased some Rode Procasters for VO work – sound great for most people, sounds terrible when I use them. Look for a mic that requires the least amount of eq in post, for the voice you are recording.
3) Always use a pop filter to take care of most plosives and if you are still getting the occasional plosive, re-position the mic so it’s slightly off axis to the mouth.
4) Make sure you are hitting your interface with the right level in 24bit. Assuming there is no outboard gear involved, make sure you are getting a healthy signal to the interface that doesn’t peak. A quick and dirty rule we use is average between -18 to -12 db with peaks no louder than -6db. 24bit gives you heaps of bits to work with in post.
Production:
5) Noise reduction – Spend some time on your recording environment so you are getting the best quality with the least amount of noise to you microphone before you do anything else. Then there’s little to no need for destructive noise reduction in an even semi-controlled environment. See point 1.
6) Normalising – Don’t normalise, just record a little hotter if you need to. If -6db is where you want your peaks, record at that volume. BTW, peaks don’t mean much at this point – a single harsh plosive could easily hit -6db with dialogue average at -24. See points 3 and 4.
So far, two steps that are unnecessary if we just spent some time on our process.
7) Work non-destructively, it’s a no-brainer. If using Audition, use the Effects Rack so you don’t commit anything until you render the final mixdown.
8) First up I add my high/low pass filters and de-essing if necessary. This is personal preference and is my workflow. I find de-essing up front allows me to high eq / presence later on without worrying about ice picks to the ears. I sometimes EQ out any ugly frequencies at this point, especially in the low/low mid area, so I don’t feed the compressor(s) these frequencies. Lows have more energy tend to make compressors work harder.
9) Compression – He gives a nice explanation in the video. For super smooth compression, I use two compressors. First compressor, catch the peaks and only the peaks with a fast attack and release and higher ratio. Second Compressor – a slow, gentler compressor, more opto in style. Use make-up gain to get the level where you want it – ie. don’t normalise.
10) EQ – Personal preference but careful boosting like he does in the video. The low end in this example would be way over the top on any speaker bigger than a mobile phone. This is the reason I lowpass filter at the beginning – it gives me more scope to boost lows for that huge sounding voiceover without sounding like an earthquake. A gently bandpass boost at around 6k for presence but be careful, easy to go over the top with this frequency and remember, if mixing for radio, they will likely have enhancers on their signal chain. Boosting will add level your VO, so watch the meters.
11) Subtle saturation – Saturation can act like an enhancer and is something we don’t naturally get in the digital realm. Analogue gear would impart subtle saturation and harmonics to the signal and give it a certain organic sound. When we added 5-6-7 different bits of analogue gear to our chain, these would add up and give that sweet analogue fuzziness that we all love. Add very subtle saturation at any point of the chain to simulate this if you want. A little salt and pepper throughout the chain really makes a difference.
10) Normalise – Don’t normalise again.
11) Only if necessary, add a limiter to tame any errant peaks.
Done.