Applies to version: 1.1.0

View the code

Go to function

IHCENVELOPE - Inner hair cell envelope extration




ihcenvelope(insig,fs,methodname) extract the envelope of an input signal insig sampled with a sampling frequency of fs Hz. The envelope extraction is performed by half-wave rectification followed by low pass filtering. This is a common model of the signal transduction of the inner hair cells.

The parameter methodname describes the kind of low pass filtering to use. The name refers to a set of papers where in this particular method has been utilized or studied. The options are

'ihc_bernstein' Compute the Hilbert envelope, compress the envelope by raising it to the power .2, combine the envelope with the original fine-structure, half-wave rectify it, square it and low-pass filter it with a cut-off frequency of 425 Hz. This method is defined in Bernstein (1999). Note that this method includes both a compression and an expansion stage.
'ihc_breebaart2001' Use a 5th order filter with a cut-off frequency of 770 Hz. This method is given in Breebaart (2001). Page 94 in Breebart's thesis.
 Filter order for the Breebaart filter, default: 5.
'ihc_dau1996' Use a 1st-order Butterworth filter with a cut-off frequency of 1000 Hz. This method has been used in all models deriving from the original 1996 model by Dau et. al. These models are mostly monaural in nature.
'ihc_king2019' Use a 1st-order Butterworth filter with a cut-off frequency of 1500 Hz.
'hilbert' Use the Hilbert envelope instead of the half-wave rectification and low pass filtering. This is not a releastic model of the inner hair envelope extraction process, but the option is included for completeness. The Hilbert envelope was first suggested for signal analysis in Gabor (1946).
'ihc_lindemann' Use a 1st order Butterworth filter with a cut-off frequency of 800 Hz. This method is defined in the Lindemann (1986a) paper.
'ihc_meddis' Use the Meddis inner hair cell model.
'minlvl' Set all values in the output equal to minlvl. This ensures that the output is non-negative and that further processing is not affected by unnaturally small values. The default value of [] means to not do this.
'dim',d Work along dimension d.


L. Bernstein, S. van de Par, and C. Trahiotis. The normalized interaural correlation: Accounting for NoSπ thresholds obtained with Gaussian and l̈ow-noisem̈asking noise. J. Acoust. Soc. Am., 106:870--876, 1999.

J. Breebaart, S. van de Par, and A. Kohlrausch. Binaural processing model based on contralateral inhibition. I. Model structure. J. Acoust. Soc. Am., 110:1074--1088, August 2001.

T. Dau, D. Pueschel, and A. Kohlrausch. A quantitative model of the effective signal processing in the auditory system. I. Model structure. J. Acoust. Soc. Am., 99(6):3615--3622, 1996a.

D. Gabor. Theory of communication. J. IEE, 93(26):429--457, 1946.

W. Lindemann. Extension of a binaural cross-correlation model by contralateral inhibition. I. Simulation of lateralization for stationary signals. J. Acoust. Soc. Am., 80:1608--1622, 1986.