Three Audio Secrets To Improve Your Audio in DaVinci Resolve

I decided to edit my NAB interviews using DaVinci Resolve this year. First, because I need the practice; second, because I want to use speech-to-text (which requires the paid version); and, third, because I really like what the Fairlight page can do to improve audio.

This became especially true after talking with Blackmagic’s audio gurus at the annual NAB Show last week. What I learned will make a big difference in the sound of my projects.

SECRET #1 – CLEAN AUDIO FOR SOCIAL MEDIA

If you only learn one secret, this is the one to learn. All my projects are posted to my website and social media channels. While you can send anything to YouTube, it will sound best if you meet their audio specs for levels and formats.

Resolve makes this SUPER easy – but almost impossible to find.

First, make your audio sound the way you want – clean up mistakes, add effects, the works – then go to the Deliver page (shortcut: Shift + 8).

At the top, select the YouTube preset – or whichever social media preset is relevant for your export.

Mid-way down the left side, check Normalize Audio.

Then, click Optimize to standard.

That quickly, Resolve will adjust levels so that it sounds the way you want AND meets all of YouTube’s specs. This is a huge time-saver and, as I listened, all my mixes were nice, clean and loud, with no distortion.

---

SECRET #2 – REDUCE BACKGROUND NOISE

When we go on location, we often want background noise. But, we don’t want it too loud. Finding this balance can be a tricky proposition.

If you have only one mic with both actor and background, you’ll need to use noise reduction to reduce the background – and that can be a hit-or-miss proposition.

But, if you have multiple mics open at once, there will be stretches where an actor, or guest in my case, is not saying anything. In this case, you want the background noise on that mic to remain, but not be as loud.

A noise gate is the answer.

Got to the Fairlight page (shortcut: Shift + 7). Double-click the Dynamic icon (red arrow) for the track to which you want to apply the noise gate.

Don’t PANIC! This isn’t as bad as it looks.

What the noise gate does is REDUCE audio that is below a level that you set. This means that you’ll still hear the audio, but it won’t be as loud as before.

• Turn on Dynamics (top left arrow)
• Turn on Expander (lower left arrow)
• Click the Gate button (lower right arrow)

Then, play the timeline where there is no talking on that track.

• Adjust Threshold so that most green dots are to the LEFT of the vertical teal-colored line.
• Adjust Range to decrease (not remove) the background noise. Generally, the default is fine.
• Leave Ratio and Attack at their default
• Set Hold and Release between 500 – 1000

Done.

---

SECRET #3 – MINIMIZE DYNAMIC RANGE IN DIALOG

Dynamic range is the “distance” in audio level between the softest and loudest spoken passages. Sometimes, like in drama, you want a lot. Other times, in interviews, documentaries and news, you don’t want much.

The multiband compressor is used to reduce dynamic range and it does it by boosting, or reducing, specific frequency ranges.

Click the Plus icon (red arrow) for the dialog track to which you want to add audio compression. Then, navigate to Dynamics > Fairlight FX > Multiband Compressor.

Holy Smokes! This is worse than the noise gate!

Rather than go into excessive detail, here are the settings I use for a male voice. The white bars at the top can be dragged left and right. This is dividing a male voice into four bands: Below human speech, low vowels, higher consonants, and above male speech.

While you can adjust anything, I recommend you match my settings for frequency ranges (drag the white vertical lines) and leave the other settings alone, except for Master Gain. Set this between 0 – 4 dB.

NOTE: Remember, final levels will be set in the Deliver page (see Secret #1).

Here are my settings for a female voice. Either of these can be saved as a preset by clicking the Plus icon in the top left corner, which is what I did for both of these.

Finally, for those tracks to which you applied the multiband compressor, change the calculation order to EQ DY FX.

This processes EQ settings first, dynamic settings second, then Effects settings last. In general, you want limiters and compressors to be last thing processed in the audio chain.

SUMMARY

Yup, my brain exploded when I learned this stuff, too. But, when I went back home and started applying these, my audio got a whole lot better. Which is why I’m sharing it with you.

---

Bookmark the permalink.