Modeling sound-source localization in sagittal planes for human listeners Seminar

Human sound-source localization in sagittal planes, including front-back discrimination, relies on monaural spectral cues. These cues result from the acoustic filtering of incoming sounds by the torso, head, and pinna. While acoustic spectral features are well-described by head-related transfer functions (HRTFs), models for sagittal-plane localization performance have received little attention so far. Hence, we developed a probabilistic, phenomenological model for human localization of stationary sounds in sagittal planes. It approximates spectral auditory processing, accounts for acoustic and non-acoustic listener-specificity, allows for predictions beyond the midsagittal plane, and directly predicts commonly used psychoacoustic performance parameters quantifying confusion rates, accuracy, and precision in localization. The validity of the listener-specific modeling approach is shown for various experimental conditions. The model successfully predicts effects on localization performance of band limitation, spectral warping, non-individualized HRTFs, spectral resolution, spectral ripples, and high-frequency attenuation in speech as reported in previous publications. Applications of the model are shown including predictions of psychophysical results and more applied results in the context of spatial audio systems or hearing-assist devices.