Audio Mastering | Measurements Related to Level | Recording Studio | Berklee Online | Jonathan Wyner

Audio Mastering | Measurements Related to Level | Recording Studio | Berklee Online | Jonathan Wyner

September 18, 2019 0 By Kailee Schamberger


Specs that relate to level are so important in the discipline of mastering. In fact, I think you could make an argument that level is the most important thing in mastering. It shows up in conversations with clients about loudness. It shows up in clients’ expectations. It shows up in questions coming from mix engineers. It shows up in thinking about output and distribution. It shows up in thinking about how the consumer, the listener is ultimately going to hear audio. If you’re not a master of anything else, understanding level and the measurements related to level is paramount. Once you have that under your fingertips, sometimes, the things that we notice about level will inform the creative choices and some of the decisions that we make in the actual mastering work. So let’s dive in a little bit to the variety of ways that we can measure level. Some of these are probably very familiar to you. Some of these are probably slightly less familiar to you, but I want to make sure that we spent a little bit of time talking about them and thinking about them together. So if you look at the screen, within ozone itself on the right-hand side of the UI, you will see that there are some faders, level controls, and there also are meters in the middle associated with it. You see the meters are active now and those meters are showing us peak level and RMS level. My assumption is that all of you understand what those are, peak level, a momentary or instantaneous measurement of level and RMS is an averaged level. That’s all very useful information. Peak especially is interesting because it tells us something about the relationship to distortion. The relationship between peak and RMS is what’s most interesting. Sometimes that’s called crest factor. It’s a fancy term to describe that relationship. But this actually tells us something about ultimately the perceived level that’s going to come out of the speakers as we’re working, and we’ll spend more time talking about that when we look at limiters and compressors. Beyond that, we’ve gotten into some more sophisticated ways of measuring level that have started to show up in distribution, in the consumer world, and the tools that are available to audio producers. In this metering environment called Insight, you will see a section or an area that’s dedicated to this idea of loudness. Let’s just define what this kind of metering is. This is something called LUFS or LKFS or LUFS. You’ve probably heard this term before. The idea of a LUFS measurement is an averaged measurement of level that is married to the idea of perception or Fletcher Munson. If you take a look at a 50 hertz tone parked at minus 20 dBFS on a standard RMS meter and you look at that same tone at 1K on an RMS meter, you’ll see that they measure the same. They both measure minus 20. But immediately you recognize that 50 hertz is going to sound way way quieter than 1,000 hertz is going to sound on that meter. An LUFS meter will reflect that. If you play a 50 hertz tone at minus 14 dBFS on an RMS meter, an LUFS meter will show it to us at minus 23, reflecting that our hearing is approximately nine dB less sensitive to that same amount of sound pressure level. So LUFS meters are next level metering if you will. Within this idea, we have three different kinds of metering, three different kinds of LUFS meters; momentary, short-term, and they’re very closely related to each other, and then integrated. I just want to spend a moment breaking those down for you as well. So when I play the audio, you’ll see that the momentary and short-term measurements are pretty closely related to each other. Momentary level remains a little bit more stable in this case, short-term is varying. They are both measuring something close to an RMS measurement. It’s actually a 400-millisecond window instead of 300 which is the classic RMS window that’s used. But the short-term measurement is looking for 400 millisecond peaks within a three-second window. So it’s looking at those peaks compared to a level that’s averaged over time, which gives us a slightly better indication of how we actually understand or perceive loudness in the context of a longer musical phrase for instance. Integrated loudness is a whole other thing while momentary and short-term measurements or something that we can use in the heat of the battle during mastering if you will. Integrated loudness is a measurement of an entire program. You may have heard of loudness normalization, the technical spec that relates to it that’s used by some of the distributors out, there is something called BS.1770. It is a way of evaluating the loudness across an entire program. It’s especially important in broadcast audio for video, but also is relevant in the context of music distribution. So integrated loudness requires that you measure across an entire three-minute song, and at the end of the three minutes, you have an integrated loudness measurement. But it’s important that you understand that it’s not something we really think about while we’re in the context of doing the mastering work or even looking at meters so much when we’re doing the mastering work, but it will become important in assessing our results. when we think about distribution. [MUSIC]