Limits of mixing background sounds to foreground sound samples in psychoacoustic laboratory experiments on noise annoyance

摘要

A variety of noise annoyance investigations are conducted as laboratory experiments. To simulate a more natural and realistic scene, it is preferable to add background ambient sound samples to the noise samples. However, possible effects of background sounds on annoyance from foreground sounds should be either minimized or quantified. A laboratory experiment was conducted, in which background ambient sound samples were mixed to foreground helicopter noise samples. In a partially balanced incomplete block design, three factors were varied: noise's sound exposure level (six categories), background sound type (five categories), and background sound pressure level (four categories). Each subject received 40 of the 78 stimuli. Annoyance to each individual 20-s stimulus was rated on the ICBEN 11-point numerical scale. It was investigated whether background sound type and/or level affected annoyance. The major predictor of annoyance was helicopter's sound exposure level. Stimuli containing background sounds with dominant events were associated with higher annoyance ratings than those with less dominant background events or with no background sounds. Increasing background level accentuated the difference between annoyance from stimuli with eventful and less eventful background sounds.