Psychoacoustically based tonality model for the evaluation of noise with tonal components

Abstract

For many years in various product assessments, tonality measurement procedures have been applied to identify and quantify prominent tonal components. The perception and evaluation of sound events containing such components has become increasingly important, e.g., in the field of vehicle acoustics for the assessment of tonality due to alternative drives, or in Information Technology (IT) devices due to hard disk drive noise. Additionally, many products include fans, e.g., IT devices, household appliances, and air-conditioning systems in buildings. These fans may emit prominent tonal sounds. The effective characterization of noise with tonal components is a challenge in acoustic and sound quality measurement. Existing methods for tonality calculation often show problems when applied to technical sounds where tonality perception is caused by different physical mechanisms like pure tones, multiple tones, narrowband noises, and even broadband noise showing very steep spectral slopes. To address this multitude of perceptual phenomena in a holistic approach, a new perceptually accurate tonality assessment method has been developed based on the hearing model of Sottek, evaluating the nonlinear and time-dependent specific loudness distributions of both tonal and broadband components via the autocorrelation function. The model has been validated by many listening tests. Its background and current state are presented.