Music searching methods based on human perception

Abstract

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.

1 - 20 . (canceled) 
     
     
         21 . A method comprising:
 receiving, from a device, a query that references a selected musical recording for which a similar recording is sought;   calculating a parameter of the selected musical recording from an analysis of music in at least a portion of the selected musical recording;   determining a scalar of the selected musical recording, the scalar being determined based on the calculated parameter and indicating an extent to which a descriptor is humanly perceptible in the selected musical recording, the determining of scalar being performed by a computer;   searching a database of reference scalars each associated with a corresponding reference musical recording to obtain a search result based on a comparison of the scalar to the reference scalars, each reference scalar being determined based on a corresponding reference parameter calculated and indicating a corresponding extent to which the descriptor is humanly perceptible in the corresponding reference musical recording; and   providing the search result to the device.   
     
     
         22 . The method of  claim 21 , wherein:
 the determining of the scalar of the selected musical recording includes multiplying the parameter by a weight factor.   
     
     
         23 . The method of  claim 22  further comprising:
 determining the weight factor based on ratings submitted by human listeners of a set of musical recordings from which the selected musical recording is absent. 
 
     
     
         24 . The method of  claim 22  further comprising:
 accessing the weight factor from a lookup table that correlates the weight factor with the calculated parameter. 
 
     
     
         25 . The method of  claim 24  further comprising:
 generating the lookup table based on ratings submitted by human listeners of a set of musical recordings from which the reference musical recordings and the selected musical recording are absent. 
 
     
     
         26 . The method of  claim 22  further comprising:
 each reference scalar is determined based on its corresponding reference parameter being multiplied by the weight factor that is multiplied to the parameter of the selected musical recording. 
 
     
     
         27 . The method of  claim 21 , wherein:
 the received query that references the selected musical recording specifies a range that corresponds to the descriptor; and   the providing of the search result is based on the scalar of the selected musical recording being within the range specified by the received query.   
     
     
         28 . The method of  claim 21 , wherein:
 the calculating of the parameter of the selected musical recording includes performing a mathematical analysis of sounds of the music in at least the portion of the selected musical recording.   
     
     
         29 . The method of  claim 21 , wherein:
 the calculated parameter represents a mathematical measurement of sounds of the music in at least the portion of the selected musical recording, the mathematical measurement being selected from a group consisting of harmonicity, loudness, dynamic range, rhythm strength, rhythm complexity, articulation, attack, note duration, tempo, and key.   
     
     
         30 . The method of  claim 21 , wherein:
 the determined scalar quantifies the extent to which the descriptor is humanly perceptible in the selected musical recording, the descriptor being a text phrase selected from a group consisting of sadness, happiness, anger, symphonicness, amount of singer contribution, melodicness, band size, countryness, metallicness, smoothness, coolness, salsa influence, reggae influence, recording quality, energy level, and danceability.   
     
     
         31 . The method of  claim 21 , wherein:
 the calculated parameter of the selected musical recording is a value of an acoustic characteristic calculated for sounds of the music in at least the portion of the selected musical recording.   
     
     
         32 . The method of  claim 21 , wherein:
 the determined scalar of the selected musical recording is a value of a musical characteristic that is perceivable by humans in the music in the selected musical recording.   
     
     
         33 . A non-transitory computer-readable medium comprising a computer program that, when executed by a computer, causes the computer to perform operations comprising:
 receiving, from a device, a query that references a selected musical recording for which a similar recording is sought;   calculating a parameter of the selected musical recording from an analysis of music in at least a portion of the selected musical recording;   determining a scalar of the selected musical recording, the scalar being determined based on the calculated parameter and indicating an extent to which a descriptor is humanly perceptible in the selected musical recording;   searching a database of reference scalars each associated with a corresponding reference musical recording to obtain a search result based on a comparison of the scalar to the reference scalars, each reference scalar being determined based on a corresponding reference parameter calculated and indicating a corresponding extent to which the descriptor is humanly perceptible in the corresponding reference musical recording; and   providing the search result to the device.   
     
     
         34 . The non-transitory computer-readable medium of  claim 33 , wherein:
 the determining of the scalar of the selected musical recording includes multiplying the parameter by a weight factor.   
     
     
         35 . The non-transitory computer-readable medium of  claim 33 , wherein:
 the received query that references the selected musical recording specifies a range that corresponds to the descriptor; and   the providing of the search result is based on the scalar of the selected musical recording being within the range specified by the received query.   
     
     
         36 . A system comprising:
 a computer; and   a computer program that, when executed by the computer, causes the computer to perform operations comprising:
 receiving, from a device, a query that references a selected musical recording for which a similar recording is sought; 
 calculating a parameter of the selected musical recording from an analysis of music in at least a portion of the selected musical recording; 
 determining a scalar of the selected musical recording, the scalar being determined based on the calculated parameter and indicating an extent to which a descriptor is humanly perceptible in the selected musical recording; 
 searching a database of reference scalars each associated with a corresponding reference musical recording to obtain a search result based on a comparison of the scalar to the reference scalars, each reference scalar being determined based on a corresponding reference parameter calculated and indicating a corresponding extent to which the descriptor is humanly perceptible in the corresponding reference musical recording; and 
 providing the search result to the device. 
   
     
     
         37 . The system of  claim 36 , wherein:
 the determining of the scalar of the selected musical recording includes multiplying the parameter by a weight factor.   
     
     
         38 . The system of  claim 37 , wherein the operations further comprise:
 determining the weight factor based on ratings submitted by human listeners of a set of musical recordings from which the selected musical recording is absent.   
     
     
         39 . The system of  claim 37 , wherein the operations further comprise:
 accessing the weight factor from a lookup table that correlates the weight factor with the calculated parameter.   
     
     
         40 . The system of  claim 36 , wherein:
 the received query that references the selected musical recording specifies a range that corresponds to the descriptor; and   the providing of the search result is based on the scalar of the selected musical recording being within the range specified by the received query.