Music searching methods based on human perception

摘要

A method for characterizing a musical recording as a set of scalar descriptors, each of which is based on human perception. A group of people listens to a large number of musical recordings and assigns to each one many scalar values, each value describing a characteristic of the music as judged by the human listeners. Typical scalar values include energy level, happiness, danceability, melodicness, tempo, and anger. Each of the pieces of music judged by the listeners is then computationally processed to extract a large number of parameters which characterize the electronic signal within the recording. Algorithms are empirically generated which correlate the extracted parameters with the judgments based on human perception to build a model for each of the scalars of human perception. These models can then be applied to other music which has not been judged by the group of listeners to give to each piece of music a set of scalar values based on human perception. The set of scalar values can be used to find other pieces that sound similar to humans or vary in a dimension of one of the scalars.

主权项

1. A method comprising:
analyzing a musical recording by performing digital signal processing to obtain a mathematical analysis of sounds recorded in the musical recording; calculating a first derivative parameter of the musical recording from the mathematical analysis of the sounds recorded in the musical recording; using a computer, determining a second derivative parameter of the musical recording based on the calculated first derivative parameter of the musical recording,
the second derivative parameter being a scalar that represents an extent to which a descriptor is humanly perceivable in music represented by the musical recording; storing the determined second derivative parameter of the musical recording as a characteristic of the musical recording; receiving a search query that references a portion of a representative recording for which a similar recording is sought; analyzing the portion of the representative recording by performing digital signal processing to obtain a further mathematical analysis of sounds recorded in the portion; calculating a further first derivative parameter of the portion from the further mathematical analysis of the sounds recorded in the portion; determining a further second derivative parameter of the portion,
the further second derivative parameter being a further scalar that represents a different extent to which the descriptor is humanly perceivable in the portion; and in response to the search query, providing a search result that references the musical recording based on a comparison of the scalar to the further scalar.