Every sound has a loudness recognized by hearing mechanism. Although loudness is a sensation measure, it is a function of sound signal properties. However, the function is not completely clear. In this paper we show that loudness determination as a function of effective mean square (RMS), peak, and average values of a sound signal is possible with an artificial neural network (ANN). We did not access to experimental data, so we produced required data using ITU-R BS.1770 model to train the network. The results show that the loudness can be simply estimated using sound signal physical features and without referring to complex hearing mechanism.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.