TAKANEN2013 - Binaural auditory model by Takanen, Santala, and Pulkki 2013

Usage

output = takanen2013(insig,fs,computationType,printFigs);
output = takanen2013(insig,fs,computationType);
output = takanen2013(insig,fs);

Input parameters

`insig`	binaural input signal for which the binaural activity map should be computed. Optionally, the output of the nonlinear cochlear model by Verhulst et. al. 2012 can be used as well
`fs`	sampling rate
`computationType`	defines the type of output provided by the model
`printFigs`	boolean value that defines whether several figures illustrating the processing steps in the model are plotted or not. As default, no figures are plotted.
`printMap`	optional boolean value describing whether the resulting activity map is plotted (by default) or not.

Output parameters

output

A structure that contains different fields depending on the input arguments.

Description

takanen2013(insig,fs,computationType) computes either the binaural activity map (if \(computationType=1\) ) or the MSO and LSO model outputs from the binaural input signal (if \(computationType=2\) ).

If \(computationType=1\) , the output structure has the following fields:

`.activityMap`	Matrix that describes in which of the six frequency ranges there is activation on a given location on the map at a specific time instant
`.colorGains`	Matrix that describes the signal level dependent gains for the different activation values on the activityMap
`.colorMtrx`	RGB color codes employed for the different frequency ranges on the binaural activity map
`.levels`	Vector specifying the left/right location

If \(computationType=2\) , the output structure has the following fields:

`.leftMso`	Output of the MSO model projecting to the left hemisphere
`.leftLso`	Output of the LSO model projecting to the left hemisphere
`.leftWbMso`	Output of the wideband MSO model projecting to the left hemisphere
`.rightMso`	Output of the MSO model projecting to the right hemisphere
`.rightLso`	Output of the LSO model projecting to the right hemisphere
`.rightWbMso`	Output of the wideband MSO model projecting to the right hemisphere

Takanen, Santala and Pulkki presented a binaural auditory model that uses the outputs of models of the medial superior olive (MSO), lateral superior olive (LSO), following count-comparison principle (von Bekesy, 1930) to project the "what" processing stream output of the model of periphery on a one-dimensional binaural activity map.

The steps involved in the computation of the binaural activity map consist of:

the given stimulus is processed with a model of periphery that consists of a nonlinear time-domain model of cochlea by Verhulst et. al. (2012) and of a model of cochlear nucleus
the binaural cues are decoded in the models of MSO, LSO and wide- band MSO from the dorsal stream output of the periphery model
the outputs of the MSO and LSO models are mapped into directions ranging from -90 to 90, and combined to form one set of "where" cues for each hemisphere
the "where" cues are used to map the "what" cues originating from the ventral stream output of the periphery models on a topographically organized binaural activity map

Requirements and installation:

Functioning model verhulst2012 (see the corresponding requirements)
Much RAM (depending on the signal length)

References:

G. von Bekesy. Zur Theorie des Hoerens. Ueber das Richtungshoeren bei einer Zeitdifferenz oder Lautstaerkeungleighheit der beiderseitigen Schalleinwirkungen. Physik. Zeitschr., pages 824-835, 857-868, 1930.

V. Pulkki and T. Hirvonen. Functional count-comparison model for binaural decoding. Acta Acustica united with Acustica, 95(5):883 - 900, Sept./Oct. 2009.

M. Takanen, O. Santala, and V. Pulkki. Visualization of functional count-comparison-based binaural auditory model output. Hearing research, 309:147-163, 2014. PMID: 24513586. [ DOI ]

M. Takanen, O. Santala, and V. Pulkki. Perceptually encoded signals and their assessment. In J. Blauert, editor, The technology of binaural listening. Springer, 2013.

S. Verhulst, T. Dau, and C. A. Shera. Nonlinear time-domain cochlear model for transient stimulation and human otoacoustic emission. J. Acoust. Soc. Am., 132(6):3842 - 3848, 2012.