Auditory filters at low-frequencies : filter shape in the range 50 Hz to 1000 Hz

Publikation: Forskning - peer reviewKonferenceartikel i proceeding

Vis graf over relationer

Prediction and assessment of low-frequency noise problems requires information about the auditory filter characteristics at low-frequencies. Unfortunately, data at low-frequencies is scarce and practically no results have been published for frequencies below 100 Hz. Extrapolation of ERB results from previous studies suggests the filter bandwidth keeps decreasing below 100 Hz, although at a relatively lower rate than at higher frequencies. Main characteristics of the auditory filter were studied from below 100 Hz up to 1000 Hz. Center frequencies evaluated were 50, 63, 125, 250, 500, and 1000 Hz. The notched-noise method was used, with the noise masker at 40 dB spectral density. A rounded exponential auditory filter model (roex(p,r)) was used to fit the masking data. Preliminary data on 1 subject is discussed. Considering the system as a whole (e.g. without removing the assumed middle-ear transfer function), the asymmetry of the auditory filter changed from steeper high-frequency slopes at 1000 Hz to steeper low-frequency slopes below 100 Hz. Increasing steepness at low-frequencies of the middle-ear high-pass filter is thought to cause this effect. The dynamic range of the auditory filter was found to steadily decrease with decreasing center frequency. Although the observed decrease in filter bandwidth with decreasing center frequency was only approximately monotonic, the preliminary data indicates the filter bandwidth does not stabilize around 100 Hz, e.g. it still decreases below this frequency as predicted by extrapolation of previous results.
OriginalsprogEngelsk
TitelProceedings of Euronoise 2009 : Action on Noise in Europe - 8th European Conference on Noise Control
Antal sider10
UdgiverInstitute of Acoustics
Udgivelsesdato2009
StatusUdgivet

Konference

KonferenceEuronoise 2009
LandStorbritannien
ByEdinburgh
Periode26-10-0928-10-09

ID: 17291796