GLASBERG2002 - Loudness model for time-variant signals

Usage

[results] = glasberg2002(inSig,fs);

Input parameters

`inSig`	monaural input signal
`fs`	sampling frequency [Hz]

Output parameters

results

output struct containing the long-term and short-term loudness

Description

Optional parameters:

`'fs',fs`	model-internal sampling frequency [Hz]; it is strongly recommended to use the default of 32000 Hz
`'flow',flow`	lowest frequency at which to evaluate the outer/middle ear transfer function
`'fhigh',fhigh`	highest frequency at which to evaluate the outer/middle ear transfer function
`'order',order`	order of the FIR filter used for deriving the outer/middle ear transfer function
`'fftLen',fftLen`	length of the fft used for preprocessing the inputSignal
`'vLimitingIndices',vLimitingIndices`
	vector [1 x 5] containing boundaries [Hz] at which to separate the inputSignal during frequency analysis
`'hannLenMs',hannLenMs`
	vector [1 x 5] containing the size [ms] of the windows used for frequency analysis of the inputSignal
`'timeStep',timeStep`	integration duration for short-term loudness [s]
`'compensationtype'`	flag; specifies the outer/middle ear transfer function. Can be 'tfOuterMiddle1997','tfOuterMiddle2007', or 'specLoud'

This code calculates the excitation patterns, the long-term and short-term loudness.

Example:

fs = 32000;
t = linspace(0,1,fs);
sig = sin(2*pi*1000*t).';
inSig = scaletodbspl(sig,100);

Note that currently fs must be 32000 Hz.

References:

B. R. Glasberg and B. C. J. Moore. A Model of Loudness Applicable to Time-Varying Sounds. J. Audio Eng. Soc, 50(5):331--342, 2002. [ http ]