A novel deep neural network model was proposed to reconstruct the head-related impulse response (HRIR) by using three dimensional anthropometric parameters. Aiming at the point physiological parameters in the HRTF database of Chinese pilots’, this paper adopts a Point-Net network composed of three convolutional layers and two hidden layers. The convolutional layer is used to extract the features of physiological parameters, and the hidden layer is used to generate hrir. The spectral distortion (SD) is adopt to quantify the difference between the measured and the reconstructed HRTF. Consequently, the proposed method performs better than the deep-neural-network based model.
Personalized HRIR Based on PointNet Network Using Anthropometric Parameters
Lect. Notes Electrical Eng.
International Conference on Man-Machine-Environment System Engineering ; 2022 ; Beijing, China October 21, 2022 - October 23, 2022
2022-08-21
6 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Hrir Image Rectification Final Report
NTIS | 1966
|HRIR image rectification Final report
NTRS | 1966
|Nimbus Hrir Equipment Diagnostic Test Program
NTIS | 1965
|