HRTF Clustering for Robust Training of a DNN for Sound Source Localization
Citation:
H. O'Dwyer, and F. Boland, HRTF Clustering for Robust Training of a DNN for Sound Source Localization, Journal Audio Engineering Society, 2022, 70, 12, 1015 - 1026Abstract:
This study shows how spherical sound source localization of binaural audio signals in the mismatched head-related transfer function (HRTF) condition can be improved by implementing HRTF clustering when usingmachine learning. A new feature set of cross-correlation function, interaural level difference, and Gammatone cepstral coefficients is introduced and shown to outperform state-of-the-art methods in vertical localization in the mismatched HRTF condition by up to 5%. By examining the performance of Deep Neural Networks trained on single HRTF sets from the CIPIC database on other HRTFs, it is shown that HRTF sets can be clustered into groups of similar HRTFs. This results in the formulation of central HRTF sets representative of their specific cluster.By training a machine learning algorithm on these central HRTFs, it is shown that a more robust algorithm can be trained capable of improving sound source localization accuracy by up to 13% in the mismatched HRTF condition. Concurrently, localization accuracy is decreased by approximately 6% in thematchedHRTF condition, which accounts for less than 9% of all test conditions. Results demonstrate that HRTF clustering can vastly improve the robustness of binaural sound source localization to unseenHRTF conditions.
Sponsor
Grant Number
Science Foundation Ireland (SFI)
13/IA/1900
Author's Homepage:
http://people.tcd.ie/fboland
Author: Boland, Francis
Type of material:
Journal ArticleSeries/Report no:
Journal Audio Engineering Society;70;
12;
Availability:
Full text availableSubject (TCD):
Creative Technologies , Digital Engagement , TelecommunicationsDOI:
https://doi.org/10.17743/jaes.2022.0051Metadata
Show full item recordLicences: