Abstract
The problem of note onset detection in musical signals is considered. The proposed solution is based on known approaches in which an onset detection function is defined on the basis of spectral characteristics of audio data. In our approach, several onset detection functions are used simultaneously to form an input vector for a multi-layer non-linear perceptron, which learns to detect onsets in the training data. This is in contrast to standard methods based on thresholding the onset detection functions with a moving average or a moving median. Our approach is also different from most of the current machine-learning-based solutions in that we explicitly use the onset detection functions as an intermediate representation, which may therefore be easily replaced with a different one, e.g., to match the characteristics of a particular audio data source. The results obtained for a database containing annotated onsets for 17 different instruments and ensembles are compared with state-of-the-art solutions.