Establishing the tonal context for musical pattern recognition

We develop a method for establishing tonal contexts of musical patterns in a musical composition. This is subsequently incorporated into a system for recognition of musical patterns. Krumhansl's (1990) key-finding algorithm is used as a basis. The sequence of maximum correlations that it outputs is smoothed with a cubic spline and is used to determine weights for perceptual and absolute pitch errors. Statistically significant maximum correlations are used to create the assigned key sequence, which is then median filtered to improve the structure of the output of the key finding algorithm.