Voiceover Recording Tech: Is Louder Better? – Tuesday Tech Tip

Is Louder better? 

There’s actually not much of a debate on that point.

To be clear, we’re not talking about that wonderful neighbor with the extreme-wattage car audio system that thumps through your body as they drive past. I’m referring to marginal increases of Loudness within a comfortable listening range. Research shows – and I’ve demonstrated this consistently in my classes – humans perceive minor increases in volume positively. Marginally louder sounds better to most people.  

Of course, we don’t want audio to burst out unexpectedly, startling everyone and making us spill our coffee. There has to be some agreement in terms of our delivery target. Right now, as far as the state of voiceover audition deliverables go in late 2022, we’re not really working with any widely accepted standard. My hope is that we can develop one. That’s what I meant last week by the phrase “Competitively Loud.”

Who is your audience?

From the point of view of the casting directors or producers, it’s better if things are arriving within an expected range of loudness. Ideally, a production company would hire an audio engineer to level balance all of the submitted auditions before playing them for review. However, given the current demands of tight project turnaround times, not to mention the added cost, that’s unlikely to occur each time.

Without that balancing step, anyone reviewing the hundreds of submissions has to keep a ready hand on the volume knob, reacting quickly to handle those who have sent auditions with volume on the outliers of the loudness range. The next audition could be inaudibly quiet or teeth-rattlingly loud. Would the listener start over? It could just be simpler just to skip forward to the next audition. That does not work in our favor.

Is “Peak” enough?

The challenge right now is that a lot of voice actors focus only on Peak values. As I’ve been discussing in the last few articles, a single random waveform can result in a high Peak value. Just using a “Normalize to -3 dB Peak” approach could still leave Loudness well below a Competitively Loud level.

One production company that focuses on character auditions has provided these guidelines:

“All audition Recording levels should hover around -18db average. Do not do any post processing (EQ/Compression/Reverb/etc…).”

– One production company’s submission guidelines

This is an interesting approach. A -18 dB recorded input target is pretty conservative, but highly useable. It certainly should reduce the number of yelled lines and battle cries getting totally blown out. One question would be whether they want actors submitting the auditions to bring each separate line to that -18 dB Peak value, or just use that value as a general target for the overall recording. In the latter case, the quiet lines would fall below that and the loudest bits might jump up a bit above in a more natural range of dynamics.

There’s clear benefit to the people who will be reviewing those auditions. All the submissions should show up in a narrower volume range which has been clearly defined. It’s surprising how rarely that occurs.

A non-scientific sampling of audition Loudness

Sampling my own character auditions delivered with those guidelines, the RMS naturally drops a bit. A random sample finds many of those in the -24 to -28 dB RMS range. That is to be expected due to the extreme dynamic range found in those character auditions. Particularly if there is a lot of conversational or quiet lines, a couple of hollers won’t bring the RMS back up to the -20 dB range at which most of my commercial auditions seem to land.

With both the Commercial or the Character/Videogame auditions, I had not actually focused on the Loudness values. What’s interesting is that there was such RMS consistency as a byproduct of my approach. In both cases, the range is a compellingly narrow set of values.  More data would help. Perhaps we need a survey on how we’re delivering our audio.

Hmmmm….That sounds like a project for next week. 

How to measure Loudness in your audio files

I had a couple questions about just how to determine Loudness. For those of you using Twisted Wave, that’s one of the statistics generated when using the “ANALYZE” function under the file menu. Adobe Audition provides Amplitude Statistics, which may be turned on under your Window menu, Izotope RX has a “Waveform Statistics” tool available, and Ocenaudio provides a Statistics tool under the Analyze menu. If you have the ACX Check tool in Audacity, that will also provide RMS.

If your recording software doesn’t easily provide that information, you could use the ACX Audiolab tool, though you do need to have an ACX profile to access that page. 

This information recently went out directly to my email community. Each week I send out a new article about using technology in the home voiceover studio, and how to balance those challenges with the need to be creative behind the microphone.
If you would like to join in to receive those emails the day they publish, please take a moment to share your contact information through this sign up form.
Thank you!