Toward a Computational Model of Perceived Spaciousness in Recorded Music
A validated computational model of perceived spaciousness in sound recordings serves to represent the subjective experience of listeners. Three dimensions of spaciousness form the basis of the model: width of source ensemble, extent of reverberation, and extent of immersion. The model is trained and tested to learn the audio parameters that contribute to these three dimensions. The resulting model predicted spaciousness 32% above that of a baseline predictor. The model can be used to show an audio engineer the value of the three dimensions in real time.
This paper costs $20 for non-members, $5 for AES members and is free for E-Library subscribers.