Often, when listening to music it may have occurred to you that some tracks seem very loud, while some are soft in parts and then there are the loud parts. Also, you may have noticed newer tracks to have a consistent loudness that sometimes seems unnatural, as all instruments seem to sound just as loud as the vocals or vice versa. You may have also noticed that when driving, you have to increase the volume of the sound system as you drive on the highway then turn it down as you slow down. Have you wondered what’s at play here? How does the level of sound against the background noise influence how we listen to music and how some techniques in audio processing leads to a better listening experience? Welcome to the world of dynamic range, how our ears perceive sound and how modern digital signal processing helps improve the overall experience.

Dynamic Range

One of the abilities of human hearing is to be able to listen to the quietest of sounds and to be able to listen to the loudest of sounds with almost the same ability. This could be a soft rustle of leaves on a summer afternoon or the deafening roar of a jet engine taking off. This is the dynamic range of human hearing: the ability to be able to discern the smallest and the slightest whisper and the loudest of sounds.

Technically, dynamic range is the difference between the loudest and quietest parts of a sound, measured in decibels (dB). It’s also known as the signal-to-noise ratio (SNR).

The unit decibel is a logarithmic measure of a ratio, that is too large to be expressed in standard metric units, such as in the case of sound pressure. For example, the sound pressure Lp could be expressed as
Lp = 20 log10 (p/po) dB

Where p is the measured sound pressure in micro-Pascals, p0 is a reference sound pressure.

So if Lp were to be measured at 120dB that would mean; that the actual sound pressure would be a million times times the reference sound pressure! Of course, with 120dB being the threshold of pain, and prolonged exposure to this level of sound would result in permanent hearing loss.

Many loudspeaker brands advertise their peak sound pressure levels to exceed 135dB, and some even claim to be able to reach 145dB, at which point it seems like it would be pointless to listen to something that loud.

Modern digital sound public address gear, is capable of upto 144dB of dynamic range in theory, with 24-bit uncompressed audio, but in practice rarely exceed 118dB even for live music concerts. Recorded digital music is usually around 90-100dB with CD grade audio peaking at about 90dB.

Other older analog formats such as cassettes and vinyls have even lower dynamic range thanks to their storage format limitations. Typically audio cassettes have about 50-75dB dynamic range with 75dB being offered in the premium chrome and metal types of cassettes. For vinyl records, even the best records cannot exceed 70dB given the physical limitations of the medium.

This is why a live performance sound louder, especially classical or jazz performances which have quiet parts and also have very loud ones.

Compression

To ensure that the quietest and the loudest parts are recorded and reproduced equally well, studios apply a technique called compression. The loudest parts above a certain threshold are “compressed” so that the difference in the quiet parts and the loud parts are now reduced. Once this is done, the music can be turned up and it sounds consistent when played back through most consumer grade equipment.

With the dynamic range reduced, the track can now be turned up, and the whole thing will sound louder. That’s because the quieter parts will be closer in volume to the louder ones. Vocals have especially wide dynamic ranges and are almost always compressed during the mixing process, and often during recording too.

One of the downsides of compression is that it would make the music sound “squashed” and artificial, which sometimes works well for certain genres such as EDM and pop, but for acoustic tracks may end up losing its natural sound.

Limiting

Another commonly encountered term in sound is limiting, which is seen in most digital signal processing equipment. Limiting is similar to compression in that both compress a signal, though the objectives are completely different.

A limiter is usually applied in the loudspeaker management section to ensure that the output signal to an amplifier is kept within a certain threshold to prevent damage to a loudspeaker.

When a signal is amplified beyond a point, the sine wave characteristics of the signal is lost especially at high output levels and this resulting distorted square wave (also called clipped wave) is highly damaging to the loudspeaker coils which is designed to handle only transient levels of high current. To prevent this from happening, the limiter is designed to kick in at a threshold well under the peak handling ability of the amplifier.

Headroom

One of the most overused and abused term in the audio industry is headroom, primarily because the term is understood so little. Headroom stands for the spare capacity in the system to accommodate an additional gain before the system goes into distortion. In most cases, this would be the published SNR or dynamic range and any signal that is within the threshold should be perfectly fine.

A good audio system should have an additional headroom of 6-8dB after the system has been tuned and gains have been settled to the maximum operating limits. This will ensure that the system is never overdriven.

In conclusion, understanding dynamic range of an audio signal is critical to creating great sounding mixes, be it for recording or for live sound. Dynamics play a critical role in the designing of audio systems for installation as well and careful consideration must be given to the components in the signal chain including sources, mixers, processors and amplifiers. With the advent of digital signal processing this has become easier to predict, model and setup but it still remains in the domain of human expertise and skill to ensure a clear, undistorted mix.

Hopefully, we have managed to throw some light on some lesser discussed concepts in the audio world and how by processing dynamics in the signal we can improve the quality of listening. As music has moved from mediums such as vinyl to tape to CD to digital, we are now able to apply some of the learnings of how human hearing works and accordingly tailor listening experiences to deliver sound as natural as possible.