Metrics including time-varying loudness models to assess the impact of sonic booms and other transient sounds

Aircraft manufacturers have proposed that it is possible to build supersonic business jets that create low amplitude sonic booms that may be less objectionable to the public than those that led to the ban of civil supersonic flight over the US. Not only do the proposed sonic boom signatures differ from traditional sonic booms in amplitude, they also differ in shape. Due to the differences between these “low” booms and traditional booms, there has been increased interest in which metrics should be used to quantify the impact of these sounds. A semantic differential experiment, conducted over high quality earphones, was designed to determine the number of independent attributes people perceived and how these attributes influence acceptability or annoyance and to investigate what metrics were correlated to these factors. The work described here is focused on the performance of various metrics in predicting participants’ ratings of impulsive sounds. It was found that statistics of and calculations derived from the output of Moore and Glasberg’s time-varying loudness model were the best predictors of participant-rated loudness, startle and annoyance. Performance of other commonly used metrics is also discussed.