THE AUDITORY MODELING TOOLBOX

Applies to version: 1.6.0

View the code

Go to function

lyon2024
The cascade of asymmetric resonators with fast-acting compression (CARFAC) model (two-capacitor version)

Usage:

[CF, decim_naps, naps, BM, ohc, agc] = lyon2024(input_waves, CF);
[CF, decim_naps, naps, BM, ohc, agc, ihc_potential] = lyon2024(input_waves, CF);

Input parameters:

input_waves Audio signal sampled at the rate of CF.fs. Size: (time x ears).
CF Structure describing the model design and state. Usually, this is the output of lyon2024_design and lyon2024_init, see exp_lyon2024 for an example.

Output parameters:

CF The updated CF structure, see lyon2024_design for more details.
decim_naps Neural activity patterns (NAPs) time-decimated by CF.seglen.
naps NAPs. Size: (time x frequency x ears).
BM Motion of the basilar membrane (BM) as the filter outputs before the inner hair cell (IHC) processing.
ohc Diagnosing internals. Meaning unclear.
agc Diagnosing internals. Meaning unclear.
ihc_potential IHC potential equivalent (unit unclear).

Description:

lyon2024(..), as known as the CARFAC v2 model, is a model of the auditory periphery, reflecting the automatic-gain-control function of the auditory system and maintaining the time-domain zero-crossings depending on the signal level. It is built as a cascade of asymmetric resonators with fast-acting compression (CARFAC). This model a modification of lyon2011:

  1. The high-pass filter (HPF) has been moved from IHC stage to the stage with the cascade of resonators.
  2. Two capacitors are used (instead of one as in lyon2011).

License

This model is licensed under the Apache License Version 2.0. Further usage details are provided in the in the AMT directory "licences".

References:

R. F. Lyon. Cascades of two-pole–two-zero asymmetric resonators are good models of peripheral auditory function. The Journal of the Acoustical Society of America, 130(6), 2011. [ http ]

R. F. Lyon, R. Schonberger, M. Slaney, M. Velimirović, and H. Yu. The carfac v2 cochlear model in matlab, numpy, and jax. arXiv preprint arXiv:2404.17490, 2024.

Build with Bootstrap