Audio meters: an introduction to audio metering

You’ll often hear the adage “use your ears, not your eyes” in the audio world. Every forum is full of this piece of advice. And while it holds up in many instances, any audio engineer working in a professional audio environment will tell you that it isn’t strictly true in every scenario.

Audio metering is one of those instances. Audio meters give engineers a way to measure and visualize different characteristics of a signal over time, including things that might not be immediately obvious from listening. Audio meters have also been developed around different standards and calibrations depending on the context in which they emerged and the problems they were designed to solve.

This article will take you from the basics through to having a decent grasp of audio metering. 

Throughout this piece, you’ll see that different kinds of meters are used for different applications. You might even need to use several at once. Monitoring several audio meters can be a headache, and that’s why we decided to solve that problem with the Meter Bridge Pro.

TRY METER BRIDGE PRO FREE NOW

What is audio metering?

Audio metering is the process of measuring and visually representing characteristics of an audio signal. Depending on the type of meter you’re using, this might mean measuring: 

• The peak level of a signal 

• Its average level over time

• Its loudness

• Its frequency content

The different metrics are useful because audio is constantly changing. A waveform can contain very short peaks that are difficult to judge by ear. Two signals with similar peak levels can have very different average levels. Audio meters give you a way of understanding what is happening inside the signal.

Audio meters were designed to account for different standards or to solve particular problems. While plugins nowadays may include several types of audio metering in one “unit,” the original outboard models were usually focused on a singular approach.

A word on decibels

The basic building block we’ll need to understand before the audio meters themselves is the humble decibel (dB).

The term dates back to Bell Telephone Laboratories, where engineers coined it around 1928 to quantify power loss and signal strength across telephone circuits. It combines the metric prefix deci- (one-tenth) with the bel, a unit named for telephone pioneer Alexander Graham Bell - the bel itself saw little practical use, since it proved too coarse a unit for most real-world measurements.

A decibel is not a measurement of loudness by itself. It is a logarithmic ratio between a measured value and a reference. This makes it useful for describing the enormous range of levels encountered in audio without having to work with a massive spectrum of numbers.

The reference changes depending on the type of measurement. In digital audio, you’ll commonly encounter dBFS, or decibels relative to full scale. In this context, 0 dBFS is the maximum representable digital level before clipping. Normal signals sit below that point.

In analog audio, you might see measurements like dBu, which is in reference to electrical voltage. VU is another measurement that describes a standardized type of level indication.

This distinction is important. A reading of “-6” on one meter does not mean the same thing as a reading of “-6” on another. Each audio meter’s reference and behavior is important to understand.

What are meter ballistics?

Audio signals change extremely quickly, so a meter doesn’t display every change in level exactly as it happens. Meters use particular response times to measure audio. This is what we’re talking about when we say “ballistics.”

Ballistics describes how quickly a meter responds to changes in a signal and how quickly its indication falls back afterward. A fast meter could focus on revealing transient activity. A slower one would show a broader picture of how the overall signal behaves.

That’s why two different meters can show you quite different readings on the same signal. It’s by design. The ballistics are an important consideration when understanding what (and how) meters are measuring audio signals.

What are important audio metering metrics?

To understand how audio metering works, we’ll need to lay some groundwork. Here are some quick-fire basics.

Loudness

Loudness describes the way we experience the strength of a sound. It is subjective, and our perception changes depending on factors like frequency, level, and duration.

That means that we had to develop more universal loudness standards. These differ depending on which standards you’re looking at. 

Modern loudness meters account for these differences using defined frequency weighting and time windows. Take the ITU-R BS.1770 standard as an example. This uses K-weighting to produce an objective loudness measurement that correlates more closely with perceived loudness than peak or RMS measurements.

BS.1770 is the standard used:

• In Europe for the EBU R128 recommendation for loudness normalization

• In the US for ATSC A/85 audio loudness for digital television

What this means is that we establish loudness standards which helps inform regional normalization practises. This is why loudness, signal level, and "volume" shouldn't be treated as interchangeable terms.

Amplitude

Amplitude describes the magnitude of an audio signal. Basically, a signal with greater amplitude has a greater electrical or digital level than one with lower amplitude.

Decibels are used to describe ratios between levels, but the reference used alongside the decibel tells you what the measurement actually represents. In digital audio, that reference is commonly full scale, giving us dBFS (decibel full scale). dBFS has a maximum defined peak level.

Peak level

Peak level is the highest instantaneous level reached by a signal over a given period of time.

Peak metering can be useful for identifying potential overloads. If the peak level of a digital signal reaches 0 dBFS, there is no more digital headroom available above it.

This is one reason almost every engineer relies on peak meters at some point in their workflow.

Here, the arrows point out the peaks in this signal which, in this case, are transients.

Average level or RMS level

Peak level is good for momentary insights into a signal’s highest points. But the way we experience loudness works over a longer period of time. If it wasn’t clear from the name, average or RMS creates a reading of a signal’s average level over a predefined period.

RMS stands for root mean square. In the context of audio and metering, RMS provides a mathematical representation of the level of a fluctuating signal over a period of time.

This video from Sweetwater gives a great overview of how RMS works:

RMS is a great way to understand the general level or density of an audio signal. A peak meter might show a series of brief peaks, while an RMS meter can give you a better sense of how much energy is consistently present.

Be careful not to confuse RMS with the term “perceived loudness.” RMS doesn’t account for the frequency-dependent sensitivity of human hearing in the way that other loudness measurement systems do.

That distinction becomes particularly important when we get to LUFS later.

The different approaches to audio metering

There are several different types of audio meters, each designed to provide a different perspective on an audio signal.

1. Peak meters

A peak meter or sample peak meter measures the instantaneous or sample peak level of an audio signal and displays the highest value reached. If you understood the above section on peak level, you probably already have a good idea what this kind of meter will do.

In digital audio, a conventional sample peak meter looks at the individual samples that make up the digital waveform and displays the highest sample value.

“True peak” meters take this concept and stretch it a bit further. These meters were created with the understanding that dBFS in the digital world is not the end of the story. 

These meters estimate the potential peaks of the reconstructed signal between individual samples. These intersample peaks can be higher than the highest individual sample value. In other words, a signal can appear to have headroom on a sample peak meter while still producing a higher reconstructed peak.

If we imagine that the white border here is 0 dBFS, intersample peaking occurs above this level, represented by the red parts of the waveform.

That’s why engineers will use true peak meters when preparing audio for distribution or converting between formats.

2. RMS meters

Once again, if you understood the RMS section above, you probably have a good idea of what to expect here. These meters measure the RMS level of an audio signal. Remember that RMS provides a mathematical representation of the signal's average level over a particular period, usually 300-600 milliseconds.

RMS can be useful when understanding or comparing general signal levels.

To wrap our heads around the practical employment of an RMS meter, let’s look at a simple example:

Imagine two signals. One (A) with spiking peaks and a low level in between those peaks. The other (B) with peaks that maintain a consistently higher level in between the peaks. What kind of readings would you get from the peak and RMS meters?

Signals A and B might have similar peak readings. The real difference would show up on an RMS meter.

The peak meter could well show a similar reading for signals A and B. The RMS level would differ significantly as it zeroes in on those longer-term readings.

Both, of course, are important. Understanding the relationship between them is key to understanding dynamics.

3. VU meters

Iconic VU meter design on a Technics RS-612US cassette deck. Image by Daniele Napolitano.

VU stands for Volume Unit. The VU meter was designed to provide an indication of the average level of an audio signal rather than reacting instantaneously to every peak.

This is done by creating a response time of 300ms. That makes it useful for observing the general level of audio material. This is the strength and limitation of the VU meter. It gives you a useful picture of average level, but fast peaks can pass by without being clearly represented.

The VU brings up an interesting discussion with regard to calibration conventions. In professional analog environments, 0 VU is often aligned with a nominal operating level like +4 dBu. But objectively speaking, 0 VU itself is a relative reference. Therefore, it can be calibrated differently depending on the system.

4. PPM meters

PPM stands for Peak Programme Meter. Unlike a VU meter, which is built to show average level, a PPM is built to protect against overload while still being legible to the eye. That’s something that a raw sample peak meter, which can flicker with every transient, struggles to do.

This is achieved through "quasi-peak" ballistics. A fast rise time lets the meter catch a transient almost immediately, while a deliberately slow decay time holds that reading long enough for an engineer to actually see it. 

You can think of this effectively as the inverse of the VU meter's symmetrical 300ms response. It’s why PPM and VU readings on the same signal can look quite different.

PPM ballistics aren't universal, though. The BBC PPM (Type IIa) uses a 10ms integration time with a 24 dB drop over 2.8 seconds, on a 1–7 scale. The DIN and Nordic PPM (Type I) use a faster 5ms integration time with a 20 dB drop over 1.7 seconds, calibrated directly in dB. The EBU PPM (Type IIb) shares the BBC's ballistics but uses its own scale calibration. All three are defined under IEC 60268-10, but a reading on one doesn't translate directly to a reading on another. The same calibration caveat that applies to VU meters applies here too.

PPM remains common in European broadcast and live sound workflows, where a fast, readable peak indication is more useful than the jumpiness of a true sample peak meter or the sluggishness of a VU meter.

5. K-System metering

Legendary mastering engineer and author of Mastering Audio: The Art and the Science, Bob Katz, developed the K-System as a way of establishing more consistent reference levels in digital audio. This system opened up different alignment options around different amounts of headroom, rather than focusing solely on 0 dBFS.

The three commonly encountered K-System scale versions are K-20, K-14, and K-12. The number refers to the nominal reference point below digital full scale, giving K-20 the greatest amount of headroom and K-12 the least.

The different versions allow you to establish different reference values for peak and RMS levels, depending on the version. K-20 is generally associated with material requiring greater dynamic range. K-14 and K-12 provide progressively less headroom. These different versions establish a reference framework for judging level and dynamics within a digital workflow.

There is a subtle workflow change if you’re using the K-System. You’ll be focusing on anchor points established by the K-Scale, rather than focusing on the absolute value of 0 dBFS. The system also requires some monitoring calibration to allow for consistency.

Setting up a K-System is a worthwhile undertaking, but it requires some work. You might want to do some additional reading.

6. The Dorrough meter

The Dorrough meter solved a problem that creator Michael Dorrough identified in the VU and PPM standards of the time.

This metering solution was designed so that engineers could simultaneously see peak and average levels. This gives engineers a better understanding of the relationship between those values. The Dorrough meter is great for understanding the dynamic structure of a signal.

The original Dorrough 40 series, recreated in Massive Meter.

The gap between the two readings can tell you a few things about your signal. A large gap means there is a significant difference between the peak and average levels. This could mean a higher crest factor and greater dynamic range. A smaller gap indicates a lower crest factor and can suggest more dynamically constrained or dense material. That’s often as a result of compression or limiting.

What is LUFS in audio?

What is LUFS?

LUFS stands for Loudness Units relative to Full Scale. You might have also heard of it being called “LKFS” or Loudness, K-weighted, relative to Full Scale. It is another standardized unit for measuring loudness. The terminology you encounter will depend on which region’s delivery specs you’ve received. Don’t worry about the name, though. LUFS and LKFS are effectively the same measurement.

Unlike a basic peak or RMS measurement, LUFS measurement takes aspects of human hearing into account. The ITU-R BS.1770 algorithm applies K-weighting to the signal before calculating its loudness. The weighting is designed to account for aspects of how we perceive different frequencies.

Basically, not all frequencies are created equal, and our ears don’t respond equally to every frequency. Our perception of loudness changes depending on frequency and level. That means that two signals with similar electrical or digital levels do not necessarily sound equally loud.

The concept is related to the equal-loudness contours used in psychoacoustics, which describe how much sound pressure is required at different frequencies for sounds to be perceived as equally loud.

This measurement approach forms the basis of practical loudness standards such as EBU R 128, which uses ITU-R BS.1770 for loudness measurement and normalization in European broadcast. EBU R 128 specifies a target programme loudness of −23 LUFS and a maximum true peak level of −1 dBTP.

LUFS doesn't attempt to perfectly replicate human perception, though. This particular measurement was designed to correlate closely with perceived loudness.

What is LUFS in audio measuring?

LUFS measurements are commonly divided into three time scales: integrated, short-term, and momentary loudness.

1. Integrated loudness: this represents the overall loudness of a piece of audio over an extended period. It is particularly useful for understanding the overall loudness of a complete program, broadcast segment, or piece of music.

2. Short-term loudness: this uses a three-second time window. This makes it useful for observing how loudness changes over relatively short sections of audio without reacting as quickly as momentary loudness.

3. Momentary loudness: similar to above, but with a 400 ms time window. This gives you a much more immediate indication of changes in loudness.

LUFS vs RMS: what's the difference?

RMS and LUFS can both give you information about the overall level of a signal, but they are measuring different things.

RMS is fundamentally a mathematical measurement of the signal's level or power over time. LUFS applies standardized frequency weighting and time integration to indicate “perceived” loudness.

This means that two signals with similar RMS levels can produce different LUFS readings, particularly when their frequency content is significantly different.

The practical difference becomes clear when looking at streaming normalization. Spotify currently uses -14 dB LUFS as its default normalization level, while Apple's Logic Pro documentation describes around -14 LUFS-I as a typical “target loudness for many streaming platforms.”

True peak is another important consideration. Spotify recommends keeping masters at or below -1 dBTP for its recommended -14 LUFS workflow, particularly to reduce the risk of distortion during lossy encoding.

Neither measurement makes the other unnecessary. They simply answer different questions. 

The world of audio metering can get very deep!

Why do engineers use multiple audio meters?

By now, it should be clear that no single meter gives you a complete picture of an audio signal. Each one provides a different piece of information, so they’re all valuable.

Think about the process of mixing or mastering. You might want to watch peak levels while also keeping an eye on average level and loudness. Looking at these measurements together can help you understand the dynamics of a signal in a way that none of them could provide individually.

Now we have a problem. How are you supposed to actually get a read on all these different meters at once?

What is a meter bridge?

A meter bridge is a centralized display that brings multiple types of metering together in one place.

The idea comes from physical studio and broadcast environments, where engineers could have a row of dedicated meters positioned together for easy monitoring. In a modern DAW, software can bring the same basic principle into one window.

Evergreen Audio’s Meter Bridge Pro lets you see all of your metering in one workspace. It’s a quick way to get a grip on the plethora of audio metering solutions we discussed in this article.

Meter Bridge allows you to monitor multiple instances of metering in one place.

Know your audio meters

Knowing the different kinds of audio metering available will make you a better audio engineer. Always remember, no one kind of meter can give you the full picture. Variety is the spice of life, so bring in more meters!

On that note, it’s worth mentioning that this article focused primarily on level and loudness metering. There are other metrics to take into account when metering, such as spectrum analysis and phase.

Get a centralized view of your loudness metering:

TRY METER BRIDGE PRO FREE NOW