One of the biggest new features in Ferrite v1.5 is Auto-Levelling. Often I make a video to show off new features of Ferrite, but with Auto-Levelling there’s not a lot to see on-screen: you just switch it on, and let it work its magic. I thought I’d write up a post about it instead, rather than trying to explain it in subtitles on what would be a mostly-static video!
Auto-Levelling is a powerful tool that automatically turns the volume levels of your project up and down, with the goal of making your production sound smooth, even and professional throughout.
(Note that Auto-Levelling is one of the paid features, available as part of the FX & Automation upgrade.)
It’s more than just an automatic gain control or dynamic range compressor — your audio passes through a process of around 10 stages, as Ferrite analyses and adjusts the results. It’s attempting to correct problems such as people moving their heads away from the mic and becoming fainter for a while, or interviews or panel discussions where multiple people have been recorded, each at different volume levels.
At the same time, it’s trying to retain the variation in level that’s natural in human speech, avoid making voices sound unnaturally flat, avoid magnifying noises that are supposed to be quiet (for example, soft consonants like “n” or “m”), and also tries not to make music fading in or out sound super weird either.
Ferrite’s Auto-Levelling incorporates aspects of normalisation, dynamic range compression, peak limiting to avoid clipping, and noise gating, but it isn’t really any of those things specifically, it’s a custom algorithm that provides many of the same benefits.
Some of these tools are already available in Ferrite. You can also use its Automation tools to adjust the volume of each track individually to your exact specifications. If you’re comfortable using those tools already, they continue to work and will give you the most control over the final result.
But Auto-Levelling takes care of the whole process with you just needing to pick a single option, so if you’re not familiar with those tools, or would just prefer not to have to deal with that stuff by hand, Auto-Levelling can save you a whole bunch of time!
To enable Auto-Levelling in a project, open it, and open up the Final Mix menu. Tap on the Auto-Levelling option, and then pick the strength of Auto-Levelling you want to apply (“Regular” is recommended in most cases):
You’ll notice that Auto-Levelling is filed under “Post-Production”, and you won’t hear any change to your audio while you are editing. This is because to achieve the results it does, Auto-Levelling needs to analyse your entire finished project from beginning to end, and use that analysis to plan its adjustments. It does this when you share/export your project.
And that’s all you need to do! I recommend making sure you have plenty of storage space free on your iPhone/iPad when you use Auto-Levelling, as it needs space to work — about 21MB per minute of finished audio. This is just temporary, of course — it will be freed up when your export is complete.
The Auto-Levelling strength adjusts how aggressive it is about changing volume levels in a short space of time — in other words, it adjusts that balance between keeping the volume level and allowing for natural variation. “Regular” is recommended for most cases, but in extreme cases — perhaps an interviewee that just would not sit still! — you may find the stronger version helpful, but it may also sound less natural as a result.
Another tip: While Auto-Levelling is normally designed to be applied “fire-and-forget”-style at the end of a project — which is the way I’d recommend for most users — if you prefer, you can also use it earlier in the process.
For example, if you have a recording you want levelled, but still want final control over how its mixed, you could use a workflow like this:
- Create a temporary project for that recording alone
- Switch on auto-levelling
- “Export” the project, but when you do, pick “Save To…” and then pick your iPad/iPhone as the destination
- This will save a levelled version of the audio back to your audio library
You can now remove the temporary project, and use the already-levelled version of the audio in a project where you leave Auto-Levelling off and adjust the levels your own way.
Either way, whether you apply it “early” or “late” in the process, I hope folks find it a useful time-saver!