Complete Communications Engineering

When designing an acoustic space, there are several subjective phenomena that can be quantified for objective measurement. By quantifying these phenomena, the acoustician can analytically model the acoustic space as it would be experienced by human listeners. The ten main subjective phenomena we are concerned with at VOCAL Technologies are Apparent Source Width, Clarity, Reverberance, Loudness, Intimacy, Warmth, Brilliance, Spaciousness, Localization, and Intelligibility.

Apparent Source Width (ASW) simply means how large the source appears to a human listener. It is defined objectively as the Lateral Energy Fraction (LEF):


Where ED is the energy of the direct sound, and ER80 is the energy of the early reflections arriving within the first 80 msec after the direct sound. When we consider only the energy after the first 80 msecs, we get the objective correlate of Clarity, the Early-to-Late Reflection Ratio (ELR):


Where ET is the total energy of the signal. Reverberance is simply the persistence of sound with a smooth decay. It can be objectively measured via the reverberation time RT60, defined as the time it takes for a source level to decreased by 60 dB:


Where V is the volume of the room, S is the absorptive surface area, ā is the average absorption coefficient for the room surfaces, and C20 is the speed of sound in air at 20 degrees Celsius. Loudness is the perceptual correlate of intensity or amplitude of a sound source. To define loudness, we use the one of the classic human loudness curves such as the dbA scale, which is simply a frequency warping of the dB scale. This A-weighting scale is defined in IEC 61672:2003:


Intimacy, also referred to as presence in some circles, is a measure of the perceived closeness of the sound source. It can be objectively assessed via evaluation of the initial time-delay gap (ITDG):


Where tRj is the time of arrival of the first reflection, and tD is the time of arrival of the direct sound. To achieve a sense of intimacy, we should have ITDG ≤ 20msec. Warmth is a feeling of a deep, rich sound, and is defined by the bass ratio (BR):


Where RT60f is the reverberation time at frequency f. To achieve warmth, we need BR > 1. Similarly, brilliance is a feeling of a light, thin sound, defined by the treble ratio (TR):


To achieve a brilliant sound, we would need TR > 1. Spaciousness is exactly as it sounds, a sense of being in a large room. To measure spaciousness, we compute the interaural cross correlation coefficients


Where f and g are the signals at our left and right ear respectively. A perfectly in phase signal will have IACC = 1, while perfectly out of phase will have IACC = -1. If they are completely different signals, IACC = 0. To achieve spaciousness, we need IACC < 0. Localization is simply the perceptual ability to distinguish the physical location of a sound source. It can be measured by the early loudness level (ELL):


In other words, it is the inverse of ASW. If a sound source is perceived as large, it will be hard to give a precise location is space. Similarly, if a sound is well localized, it will not be perceived as having width.

Finally, intelligibility can be defined several ways. The first way is well known Speech Transmission Index, which is a complex procedure aiming at measuring the sound transmission ability of a room. Its practical implementation is called RASTI (Rapid STI), defined in IEC-60268-16. In 2011, RASTI was declared obsolete and replaced with STIPA (STI for Public Address systems) in IEC-60268-16.4. In addition to STI, one can measure %Alcons, which is the percentage articulation loss of consonants, another complex measurement procedure measuring the ability of a listener to understand consonants.

At VOCAL Technologies, we have developed extensive modeling tools to exactly reproduce both your problematic and desired acoustic environment, in part by intelligent implementation of the above metrics, thereby allowing us to tailor a speech quality enhancement solution to your exact needs. Contact us today to discuss your application and your environment, and we will provide a unique solution that is sure to alleviate your speech quality problems. For more details on our speech quality enhancement solutions, continue browsing our site.