June 19, 2017

FML QuickBrief - June 2017

Prepping Your Content for Processing

 

The processing core that drives FixMyLevels is pretty powerful, but it can't make useful corrections if the source material is not properly mixed from the start. This is especially important for podcasts that feature guests speaking at different levels. If the host is coming in hot and the guests are 5-10dB lower, the FML gain control algorithm is going to be constantly chasing the target loudness level. For your listeners, this is going to create the effect of someone rocking the volume control up and down.

 

Ideally, you'll want to try to get this right from the start. That means running a sound check with your guests ahead of the actual recording. Have them speak into their respective microphones at a conversational level. Use your eyes and your ears to make sure everyone is coming in within a dB or two of each other. Set the levels with enough headroom to allow for laughing, shouting, or any other unexpected gain changes that would cause clipping.

 

If you can record each voice on a separate track, that is even better! This will allow a lot of control in post production to make any final touches before creating the final mixdown.

 

BONUS: If you are using Skype to record guests, lower their overall gain before mixing down. The host, who is likely speaking into a somewhat decent mic, is going to have a much lower perceived loudness than a guest coming in over Skype at the same gain level. Why? Because Skype is doing its own compression and roll-off to maintain quality, and the spectral imbalance between the studio mics and Skype will be enough to create a variance in perceived loudness between host and guest.

 

Individual Tracks vs Final Mix: Which One Should I Process?

 

You should only process the FINAL MIX through FML. There are a couple of reasons for this.

 

First, the FML processing core functions properly only when it can see all audio at one time. Your listeners are not receiving your podcast one track at a time, so it wouldn't make sense to only process one track at a time. Remember that FixMyLevels is not an editing tool, it is a mastering tool. It needs to see the whole picture to make intelligent compression and gain control decisions.

 

Second, you run the risk of artificially boosting the gain of your final mix if you process each track individually. That includes bringing up the noise floor of each microphone, which is a big no-no. You'll find that stacking a bunch of individually compressed audio files on top of one another will enhance the sonic energy of the final mix and cause your podcast to sound, for lack of a better term...weird.

 

Dynamic Range Is the Enemy

 

The push for loudness standards in the podcasting world is really important. All broadcasting mediums have a standard target that they aim for to maintain compliance. In podcasting, we've all heard that we need to shoot for -16 LUFS. What they don't tell you is that you can hit -16 LUFS and still drive your listeners insane with high dynamic range fluctuations.

 

Podcasts are being consumed in a world where people can download entire movies to their handheld devices and take them wherever they wish. It also means that your listeners are probably consuming your content in an environment that does not lend itself to quiet enjoyment. Personally speaking, I listen to most podcasts on an airplane. I tend to sit in an exit row when I fly, and conveniently, there is a large engine sitting about six feet from me. I do not want to listen to anything that requires me to keep my phone handy so I can keep the audio in check. Is the loudness target of -16 LUFS important? Sure. Is low dynamic range even more important from a listener's perspective? ABSOLUTELY.

 

FixMyLevels kills annoying changes in dynamic range. In fact, our processing algorithms take priority in doing this before hitting the loudness target of -16 LUFS. Why? Because listeners are not measuring the quality of your podcast with a meter that indicates your integrated loudness (in fact, they probably don't even know what LUFS are!). But they do understand dynamic range, and they will not hesitate to listen to something else if they can't get through your podcast frustration-free.

 

That's it for this month! Enjoy the beta and keep the feedback coming so we can make FixMyLevels even cooler than it is now!

1 / 1

Please reload

© 2018 FixMyLevels.com | Audio Manipulation Group, LLC

Questions?
Comments?
Snide Remarks?
support@fixmylevels.com
  • Facebook B&W
  • Twitter B&W
  • LinkedIn B&W