HRTF Clustering for Robust Training of a DNN for Sound Source Localization

This study shows how spherical sound source localization of binaural audio signals in the mismatchedhead-relatedtransferfunction(HRTF)conditioncanbeimprovedbyimplementing HRTF clustering when using machine 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 representativeoftheirspecificcluster.BytrainingamachinelearningalgorithmonthesecentralHRTFs,itisshownthatamorerobustalgorithmcanbetrainedcapableofimprovingsound sourcelocalizationaccuracybyupto13%inthemismatchedHRTFcondition.Concurrently,localizationaccuracyisdecreasedbyapproximately6%inthematchedHRTFcondition,which accountsforlessthan9%ofalltestconditions.ResultsdemonstratethatHRTFclusteringcanvastlyimprovetherobustnessofbinauralsoundsourcelocalizationtounseenHRTFconditions.