We propose a new sound onset detection method using principles of psychoacoustics. Our algorithm mimics the human auditory system to detect onsets in polyphonic audio.A filterbank based processing scheme is used that allows for a more robust detection of onsets using a popular model of loudness. We show the versatility of our algorithm by evaluating itpsilas performance over a wide range of sounds.
No takes yet. Share an insight, caveat, or question.
Thoshkahna et al. (2008) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: