China
Africa
Explore chapters and articles related to this topic
Loudness
Published in Stanley A. Gelfand, Hearing, 2017
The loudness calculation procedure itself should be performed using computer software that is readily available (Glasberg and Moore, 2007; ANSI S3.4-2007) because it is quite laborious when done manually. We will not attempt to outline the details, but the general framework is as follows. The spectrum reaching the cochlea is determined by adjusting the stimulus spectrum to account for the transfer functions (a) from the sound field* or the earphones to the eardrum, and (b) through the middle ear (see Chapter 3). From this, the excitation level in the cochlea is determined for each auditory filter, expressed as the equivalent rectangular bandwidth (ERBs, see Chapter 10),* which in turn is converted into loudness values in sones per ERB. The overall loudness of the sound in sones is then calculated by summing the values for all of the ERBs. Interested students will find an informative review of each stage in Marozeau (2011). In addition, one should be aware that the Moore model continues to evolve with the unfolding of new information and applications (for a review, see Moore, 2014). For example, modifications have been introduced when for people who have hearing impairments (Moore et al., 2014).
Acoustic analysis of the effect of personal protective equipment on speech understanding: lessons for clinical environments
Published in International Journal of Audiology, 2023
Robert W. J. Mcleod, Maria Gallagher, Andy Hall, Sarah P. Bant, John F. Culling
In order to visualise the perceptual effect of PPE, the differences in transfer function were smoothed in the fashion of cochlear excitation patterns (Moore and Glasberg 1983). This converts the difference in sound transmission with and without PPE into the change would be perceived by a listener as a function of frequency. The overall practical effect of the attenuation was evaluated using a weighting function from the articulation index (ANSI 1997, Table 1). This function weights each frequency band according to its importance in speech perception to produce a predicted reduction in the effective signal level for speech reception caused by the mask. When listening in noise there will, therefore, be a corresponding reduction in effective SNR ratio. For this purpose, these weightings were redistributed onto equivalent rectangular bandwidth (ERB)-spaced frequency bands (Table 1) using Moore and Glasberg’s (1983) Equation (5). It provided an objective and comparable measurement converting acoustic transmission into the perceived effect of the difference in sound transmission on listener experience, establishing the likely practical effect on verbal communication.
Hearing aid delay in open-fit devices – coloration-pitch discrimination in normal-hearing and hearing-impaired
Published in International Journal of Audiology, 2023
Performance of the simulations was evaluated in equivalent rectangular bandwidths (Moore and Glasberg 1983), and the error defined as the absolute difference between the mean of the prediction and the mean of the reference for those frequencies within the equivalent rectangular bandwidth. The frequency resolution for the ERB error computation was 0.032 kHz. Compared to computing the level difference in narrower bands, this effectively smoothed the errors that are not perceptually relevant, because these would not be resolved in the auditory filters. In total 96 conditions were measured (four IGs, six azimuth angles, two delays, two ears). For the two different delay settings Figure 3 shows the mean absolute ERB error (ME) and the standard deviation (SD) across the 12 conditions (six angles and two sides). An average ERB error (
Effect of age, test frequency and level on thresholds for the TEN(HL) test for people with normal hearing
Published in International Journal of Audiology, 2020
The threshold equalising noise (TEN) test was designed for the rapid diagnosis of DRs in clinical practice (Moore et al. 2000). It requires measurement of the detection threshold for a sinusoid in a special noise (TEN) that is designed to produce equal masked thresholds for all signal frequencies within a certain range, for people with normal hearing. The TEN level is usually specified as the level in a 1-ERBN-wide band centred at 1 kHz, where ERBN stands for the equivalent rectangular bandwidth of the auditory filter at a moderate sound level for listeners with normal hearing (Glasberg and Moore 1990). The most commonly used version of the TEN test, called the TEN(HL) test (Moore, Glasberg, and Stone 2004), uses levels that are specified in dB HL. The signal-to-TEN ratio at the threshold, in dB, is denoted the STR. If there is no DR at the place tuned to the signal frequency, then the signal is detected through an auditory filter centred close to the signal frequency and the STR is close to 0 dB for people with normal hearing or slightly above 0 dB for people with hearing loss. If there is a DR in the cochlea at the place tuned to the signal frequency, the signal is detected if it produces sufficient vibration at a remote place in the cochlea that is still functioning, in other words, it is detected via off-frequency listening. This leads to an STR that is markedly higher than 0 dB. The usual criteria for diagnosing a DR are that the STR should be 10 dB or more and the signal level at threshold should be at least 10 dB above the absolute threshold (Moore et al. 2000; Moore, Glasberg, and Stone 2004; Pepler et al. 2014).