Loudness Units: Phon and Sone

Q two Two units have frequently been used to measure the magnitude of loudness, phon, and sone. Please give an explanation for their meanings and the detailed procedure of how to measure them.
Phon, as a unit used for the size of the magnitude of loudness, indicates a person's perception of loudness, and 1 phon equals 1 decibel at one thousand Hz (1 kHz). The decibel scale is a logarithmic scale used to express sound intensity due to the abnormally huge range of sound intensities over which people can hear (Oberfeld, 2007). Sound depth refers to the rate at which sound energy reaches a unique cross-sectional area. The decibel rating can get determined thru the measurement of the level of sound intensity of a values provide an objective sound measure (Oberfeld, 2007). Additionally, while sound intensities and decibels (dB) are measurable quantities, the sound loudness is subjective and varies from one person to another. Besides, sounds with different frequencies but equal intensities get perceived by the same person as having unequal loudness. For example, a sound of 60 dB and a frequency 1000 Hz will sound louder than one of 60 dB and frequency of 500 Hz (Goldstein, 2013).

Therefore, two sounds with equal intensities do not necessarily have equal loudness. That is because the human hearing sensitivity varies depending on sound frequency, and a plot of equal loudness curves is helpful in showing the variations for the average human ear (Goldstein, 2013). By sing 1000 Hz as the standard frequency, it is possible to reference each equal loudness curve to the decibel level at a frequency of 1000 Hz, and that forms the basis for loudness measurement in phons. In other words, if the loudness of a particular sound is perceived to be 60 dB at 1000 Hz, then such a sound is considered to have a loudness of 60 phones. 60 phons, therefore, implies "as loud as a 60 dB, and a 1000 Hz tone (Goldstein, 2013).”

The measurement of loudness for complex sounds is achievable by comparison to 1000 Hz test tones, usually for research purposes. However, the measurement of practical sound level involves the use of filter contours. The use of phons for the measurement of sound loudness entails using several equal loudness curves, which get determined experimentally (Oberfeld, 2007). The experimental procedure for the determination of the equal loudness curves involves, first, subjecting individuals to a 1 kHz sound at 60 dB (loudness of 60 phons). Sound with different frequencies then gets played with the adjustment of the decibel level unit it gets perceived to have similar loudness to the one at 1000 Hz (Oberfeld, 2007). The procedure then gets repeated for different frequencies to generate a full 60-phon curve. The procedure can then be repeated for various decibel levels to give a result like the one shown in Figure 1 below.

On the other hand, a sone is a unit used for the measurement of sound loudness through the use of a linear sone-scale. When using sone in the measurement of sound loudness, doubling the perceived sound loudness doubles the corresponding sone value (Ozawa, Suzuki, & Sone, 2001). The sone scale is a third scale that relates to sound loudness and bases on the observation that increasing the sound level by 10 phons is perceived as loudness doubling. One sone, according to the sone scale, refers to a sound with 40 phons loudness as shown in Figure 2 below (Ozawa, Suzuki, & Sone, 2001).

Figure 2: The Sone Scale

The sone scale got created through the use of the rule of thumb for loudness to provide a linear scale of loudness. The creation of the sone scale took into consideration the assumption that the standard range for instrumental music is approximately 40 to 100 phons (Oberfeld, 2007). Therefore, if the lower limit of that range gets assigned a loudness of 1 sone arbitrarily, then fifty phons would have a loudness of two sones, and sixty phons would be four sones, and the order would continue as shown in Figure 2 above.

However, sone scale is not proportional to the phon scale. Instead, the loudness of sound, when measured in sones is, to a greater extent, a power law function of the intensity of the sound signal, with an exponent of about 0.3 (Ozawa, Suzuki, & Sone, 2001). It is due to that exponent that an increase of 10 phons (10 dB at 1000 Hz) produces sound loudness that is almost double the loudness in sones. At sound frequencies, different from 1000 Hz, the level of sound loudness in phons gets calibrated based on the frequency response of human hearing through the use of equal loudness contours. The sound loudness in phons then gets mapped to sound loudness in sones through the same power law (Ozawa, Suzuki, & Sone, 2001).

Q 3. Hearing-impaired listeners generally have difficulty in perceiving sounds in noise. Please explain at least two major reasons for their difficulty.

One of the reasons why hearing- impaired listeners have difficulty in perceiving sounds in noise is reduced sensitivity. There exists a threshold of sound detection for pure tones, below which individuals cannot perceive sounds. The threshold sound detection curve for hearing-impaired individuals shows that the prevalence of hearing impairment high in areas with sounds of high frequencies (Emiroglu & Kollmeier, 2008). The typical loss of audibility among the hearing-impaired listeners can make high frequency sounds imperceptible, hence the difficulty in perceiving sounds in noise. Additionally, since most consonants are usually at higher sound frequencies and lower loudness levels than the vowels, and since consonant sounds convey most of the information in any speech, the hearing-impaired listeners have difficulty in perceiving sounds in noise as shown in figure 3 below (Emiroglu & Kollmeier, 2008).

Figure 3: Pure-tone audibility threshold for typical hearing-impaired listener (dotted line) and normal hearing (Continuous line)

Another reason why hearing- impaired listeners have difficulty in perceiving sounds in noise is the nonlinear loudness shift caused by hearing impairment. In a psychoacoustic test, a normal and a hearing-impaired listener get presented with narrowband noise or tone at varying sound pressure levels then asked to use a predefined scale to judge the sound loudness (Apoux, Grouzet, & Lorenzi, 2001). Both the listeners (hearing-impaired and normal) usually judge high-level sounds as having similar loudness. However, the hearing-impaired listener judges the sounds at lower levels as softer than the normally-hearing listener. Such a nonlinear shift in sound loudness varies with sound loudness and frequency and indicates the sophistication level necessary to restore hearing-impaired listeners' auditory perception. Figure 4 below shows the abnormal increase of sound loudness for a hearing-impaired listener, where sensitivity to low-level sounds gets altered significantly, while that for high-level sounds is almost normal (Apoux, Grouzet, & Lorenzi, 2001).

Figure 4: Perceived tone loudness showing the abnormal loudness growth for the hearing-impaired listener, where sensitivity to low-level sounds gets altered significantly while that of high-level sounds is almost normal

The variation of sound perception in noise between hearing-impaired and normally hearing listeners is also as a result of the bandwidth of the sound signal. The sound loudness, for a normally-hearing listener, increases as the sound signal’s bandwidth increases, even when the signal power remains constant, a process termed as loudness summation (Emiroglu & Kollmeier, 2008). However, for a hearing-impaired listener, the rise in loudness as signal bandwidth increases does not take place, and the loudness remains constant irrespective of the bandwidth as shown in figure 5 below. What causes such a difference is the manner in which sound gets filtered into different channels along the basilar membrane within the ear’s cochlea (Emiroglu & Kollmeier, 2008).

Figure 5: Level of a 1000 Hz sinusoid similar in loudness to a 2-tone complex as a function of complex’s bandwidth, Demonstrating the hearing-impaired listener’s lack of loudness summation

Another cause of the hearing-impaired listeners’ difficulty in perceiving sounds in noise is the frequency resolution. The ear’s ability to resolve sound components of various frequencies is an essential aspect of hearing and sound perception in general. When one’s capability to resolve the frequency of sound gets compromised, then it becomes harder to identify sounds, thereby impairing the ability to perceive or understand speech, especially in noise (Sang, Hu, Zheng, Li, Lutman, & Bleeck, 2014). The background noise worsens the ability of the hearing-impaired listeners to perceive sound since the ear’s ability to separate the background noise from the target sound gets reduced significantly. Figure 6 below shows how frequency resolution gets affected by hearing impairment (Sang, Hu, Zheng, Li, Lutman, & Bleeck, 2014).

Figure 6: Pure tone thresholds using a 1 kHz masker, demonstrating the lowered frequency resolution of hearing-impaired listeners.

The curves in Figure 6 plot the threshold of detection for a sound tone in the presence of a 1 kHz masker and the widening of the high thresholds shows that the hearing-impaired listener has less ability to resolve the target sound from the masking tone due to the widening of the auditory filters of the hearing-impaired listener. Both the broader and narrower filter bandwidths do not differentiate various frequency regions, and the result is the sound representation in a smeared spectral manner in the system of the hearing-impaired listener (Sang, Hu, Zheng, Li, Lutman, & Bleeck, 2014).

Temporal resolution is another reason why hearing-impaired listeners have difficulty in perceiving sounds in noise. The ear's cochlea is a dynamic system that functions in a time-varying and a nonlinear manner on the auditory signal (Emiroglu & Kollmeier, 2008). The specific dynamics of the cochlea affect the manner in which sounds get perceived, as well as the people's ability to perceive such sounds over a competing background noise, specifically among the hearing-impaired listeners. An impaired auditory system distorts the temporal envelope of sound that codes essential information, thus resulting in a distorted sound perception in background noise (Emiroglu & Kollmeier, 2008).

References

Apoux, F., Crouzet, O., & Lorenzi, C. (2001). Temporal envelope expansion of speech in noise for normal-hearing and hearing-impaired listeners: effects on identification performance and response times. Hearing Research, 153(1-2), 123-131. http://dx.doi.org/10.1016/s0378-5955(00)00265-3

Emiroglu, S., & Kollmeier, B. (2008). Timbre discrimination in normal-hearing and hearing-impaired listeners under different noise conditions. Brain Research, 1220, 199-207. http://dx.doi.org/10.1016/j.brainres.2007.08.067

Goldstein, J. (2013). A phon loudness model quantifying middle ear and cochlear sound compression. The Journal Of The Acoustical Society Of America, 134(5), 4228-4228. http://dx.doi.org/10.1121/1.4831528

Oberfeld, D. (2007). Loudness changes induced by a proximal sound: Loudness enhancement, loudness recalibration, or both?. The Journal Of The Acoustical Society Of America, 121(4), 2137-2148. http://dx.doi.org/10.1121/1.2710433

OZAWA, K., SUZUKI, Y., & SONE, T. (2001). Sound Quality of Two-tone Complex Sounds with Different Overall Loudness. Interdisciplinary Information Sciences, 7(2), 237-246. http://dx.doi.org/10.4036/iis.2001.237

Sang, J., Hu, H., Zheng, C., Li, G., Lutman, M., & Bleeck, S. (2014). Evaluation of the sparse coding shrinkage noise reduction algorithm in normal-hearing and hearing-impaired listeners. Hearing Research, 310, 36-47. http://dx.doi.org/10.1016/j.heares.2014.01.006