10 Steps to Great Audio

UPDATE : Some people came to this post looking for BEGINNING setup steps for recording with a microphone. This isn’t that. If that’s what you’re here for, then I highly recommend you watch this YouTube video from Podcastage instead. You don’t want great audio yet, start with getting not-so-bad audio, progress to consistently good audio, and then this post will make sense to you.

After my recent post about Debunking The Gain Myth I realized that my approach to all things audio/video is pretty backwards to how it’s usually done, and I might want to say why.

I studied both audio production and video production, y’know. But not just in college. With my a Grammy-winning producer for a father, I literally grew up around it, and I’ve continued to experiment constantly.

First and foremost, I’m pragmatic guy. I love science and concepts, but for anything artistic — perception always wins! I’m annoyingly evidence-based about anything based in what we see or hear. I think the world would sound better if more audio engineers approached recording like this.

As a vocal percussionist / beatboxer who is intimately familiar with how specific microphones distort and how to use this to my advantage as a performer / recording engineer / sound designer / producer myself, I consider myself uniquely qualified in how to “play the microphone”. That’s why I tend to run my gain super high, at least according to common wisdom.

So basically, I do all the things you’re not supposed to do with a voice, and all the things you’re not supposed to do with a microphone. And here I am, 30 years into making weird noises that vocal coaches warned me would fry my voice and give me nodes within a year or two. Yet my voice is still healthy! And after playing with microphones for at least as long, I know how to make them work for me without damaging them just fine, and getting some award-winning recordings out of it.

Give me a microphone I’ve never seen before and 5-10 minutes, and I’ll be able to tell you all it’s strengths and weaknesses, including the frequency- specific polar pattern responses, and an estimated SPF rating. I don’t need the factory documentation, I’ve come to trust my own pragmatic calibration levels even more.

Does that sound too bold? Well, keep in mind that I don’t care what this model of microphone is supposed to do, because I can only use what this specific microphone will actually do. Maybe it got dropped or got some gunk in it or never quite lived up to spec in the first place, I don’t know. I just know what I hear.

And then there’s the rest of the signal flow after that. Which brings me to…

Sam’s Ten Steps to Great Audio (microphone-centric)

1. Do something with air that’s worth capturing.
Never try to “fix it in post” with audio, as frustration and budgetary overages are the inevitable result. Get the tone, the timing, the feel, that indescribable element, get it all right before approaching the microphone (y’know, like practice it). This doesn’t mean the performance has to be perfect, just that we have to treat whatever is there as if it is perfect at every subsequent step. So this step is where you either have perfection or give up on it and go for good enough (which is almost always exactly that!).

2. Place your microphone someplace where it gets the most desirable variances of that air.
Your ears are what should guide this process, everything you know or think you know in terms of theory is just the beginning (Cindy Crawford Mole Position is always the best starting place for vocals, btw). Experiment and make sure to think 3-dimensionally when you do. Validate to confirm that the sounds you want are being captured, and that the sounds you don’t want are not. Hear the room through the mic(s) and change the mic placement and room characteristics as necessary to get what you want. Generally speaking, technology is not the solution to this problem, listening is. So resist the urge to go buy a new microphone to fix anything, and learn to get a good result out of what you have on hand first.

3. Crank your gain up as high as possible to give the most signal from that microphone without distorting.
If you feel the urge to argue, please watch this :)

4. EQ any individual track toward the final mix.
You don’t need to wait to make space for everything else, so carve it out early. Think of a rainbow, one continuous band of refraction but made up of distinct bands of color. You can imagine each track as a color, occupying a certain band of sonic space. The rainbow is the overall mix, and it’s not important to that people can tell where one color stops and the next one starts. What’s important is that it is beautiful and full in total as itself.

5. Work with the raw & dynamic audio for as long as possible in your workflow before applying any effects.
Please don’t write me off as an a cappella purist, getting familiar with your raw materials and learning to love them for what they are will help you work better with the audio you have, as opposed to the audio you wish you had. This is one of the things that separates the good engineers from the great ones.

6. When editing, get the timing & feel right to your ears , not your eyes. Even what looks right on the screen won’t sound right if it isn’t right. Remember that the people who matter will only ever hear the output of your audio work, and timing is everything — yes, even with a simple voiceover. Again, note that in my somewhat backwards way of doing things, editing comes before applying any effects. But for that to really work well, you’ll need to know a bit about about ADSR envelopes, what effects you will apply, and what they will do for you before you get all nitpicky and slice everything up manually.

7. When applying effects, start with noise gates & compression.
This is another backwards thing I do that works very well for me, and it can for you too. People argue with me about this on occasion, but to me there’s no use mixing with artifacts that won’t stay there in the end. Just watch the cutoffs with any reverb so you don’t reverberate the exits. If you don’t know what I’m talking about, there are a million or so tutorials on YouTube to explain, here’s an oldie but goodie.

8. Mix all the way to the edges of the audible spectrum , from 20 Hz to 20,000 kHz (and even beyond), but don’t try to hit all the bands at once.
Get all 4-dimensional on it instead. Back to the rainbow analogy, the entire rainbow does not need to appear all at once to be a rainbow. And like rainbows, there are a lot more colors present than the ones that we know how to see. Depending on the project, I often even mix with sounds that people can’t technically consciously hear do mysteriously seem to impact our perception. I don’t know why this is, but from my own personal experimentation, it still seems to be true whether anyone can explain to me why or not. (My unproven suspicion is that frequencies outside of our hearing range influence how we perceive timbre, but that’s a very geeky conversation for another time)

9. Leave mastering to the Masters.
My rule is that if you recorded it, you are automatically disqualified from any opinions beyond your own final mix, so let it go. Anytime I’ve let this rule be broken still hurts my ears to this day.

10. Always remember, if something doesn’t sound good, then it isn’t good.
Never fear starting over from the very beginning with another iteration. The sooner in this workflow that you do this, the less painful/wasteful it is.