MODELLING NARROW-BAND SOUND LOCALISATION IN THE MEDIAN PLANE USING SPECTRAL GRADIENTS

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference article in proceedingsScientificvertaisarvioitu

Abstrakti

Localisation of narrow-band sounds is heavily influenced by their frequency content. Middlebrooks [1] demonstrated the relation between the perceived location of a sound and the subjects' directional transfer function. Since then, neurophysiological studies have shown evidence of the sensitivity of the dorsal cochlear nucleus to the positive gradient of the sound spectrum, offering a physiologically-inspired alternative for modelling auditory localisation. In an attempt at connecting narrow-band localisation data to the modern broad-band localisation auditory models, the current work presents recent perceptual results, which have been analysed using the positive gradient of the spectral profile. Results obtained by the spectral gradient analysis were similar to the ones obtained using the directional transfer function. Thus, using the positive gradient of the spectrum as a proxy for the internal representation of the spectral cues allows the modelling of both narrow-band and broad-band auditory localisation data with the same computational approach.

AlkuperäiskieliEnglanti
OtsikkoForum Acusticum 2023 - 10th Convention of the European Acoustics Association, EAA 2023
KustantajaEuropean Acoustics Association
Sivumäärä8
ISBN (elektroninen)9788888942674
DOI - pysyväislinkit
TilaJulkaistu - 2023
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaForum Acusticum - Torino, Italia
Kesto: 11 syysk. 202315 syysk. 2023
Konferenssinumero: 10

Julkaisusarja

NimiProceedings of Forum Acusticum
ISSN (painettu)2221-3767

Conference

ConferenceForum Acusticum
Maa/AlueItalia
KaupunkiTorino
Ajanjakso11/09/202315/09/2023
MuuConvention of the European Acoustics Association

Sormenjälki

Sukella tutkimusaiheisiin 'MODELLING NARROW-BAND SOUND LOCALISATION IN THE MEDIAN PLANE USING SPECTRAL GRADIENTS'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä