In this paper we propose a neural network-based approach for audio equalization inside a car cabin. We consider the Generative Adversarial approach to generate FIR filter coefficients for binaural equalization of the sound produced by multiple loudspeakers at the driver listening position. The neural network is optimized for the generation of filter coefficients for a flat frequency response equalization in one control position in a time-invariant scenario. Results are analyzed in the frequency domain, comparing the achieved frequency response with the desired one. Compared to previous works, the proposed approach reaches the target response with a very low error.
Authors: Giovanni Pepe (Università Politecnica Delle Marche, ASK Industries Spa), Leonardo Gabrielli (Università Politecnica delle Marche), Stefano Squartini (Università Politecnica delle Marche), Luca Cattani (ASK Industries Spa) and Carlo Tripodi (ASK Industries Spa)