Sheet Music Search and Discovery System

Abstract

A sheet music search and discovery system is disclosed that uses specific mathematical rules to analyze and characterize sheet music and provides functionality for users to identify sheet music based on those characterizations. The system stores sheet music data and metadata characterizing each composition, provides a graphical user interface that provides functionality for users to search the sheet music data for compositions, and generates search results based at least in part on the metadata characterizing each composition. In one embodiment, metadata describing structured sheet music data is generated using a global vector space that includes semantic representations of elements extracted from a large corpus. In another embodiment, metadata describing unstructured sheet music data is generated using machine learning-based pattern recognition. In another embodiment, the interface provides functionality for users to identify instruments and a range for each of the instruments and identify compositions with similar instruments and ranges.

1 .- 20 . (canceled) 
     
     
         21 . A system, comprising:
 non-transitory computer readable storage media that stores:
 sheet music data comprising compositions; and 
 metadata characterizing each composition, including one or more instruments and a range for each of the one or more instruments; 
   a graphical user interface that provides functionality for a user to identify a user profile that includes one or more instruments and a range for each of the one or more instruments;   a search engine, executed by a hardware processor, that:
 identifies compositions by comparing the one or more instruments and the one or more ranges included in the user profile with the one or more instruments and the one or more ranges of each of the compositions; and 
 displays the identified compositions to the user via the graphical user interface. 
   
     
     
         22 . The system of  claim 21 , wherein the search engine compares the one or more instruments and the one or more ranges included in the user profile with the one or more instruments and the one or more ranges of each of the compositions using Manhattan (L1) distance or Jaccard similarity. 
     
     
         23 . The system of  claim 21 , wherein the search engine identifies compositions that include one or more parts corresponding to the one or more instruments, where each of the one or more parts are within the range of the instrument corresponding to that part. 
     
     
         24 . The system of  claim 21 , wherein:
 the graphical user interface further provides functionality for the user to search the compositions; and   the search engine:
 generates a query based on the user search; 
 queries the compositions and the metadata characterizing each composition using the query to generate search results; and 
 returns the search results to the user via the graphical user interface. 
   
     
     
         25 . The system of  claim 21 , wherein:
 the graphical user interface further provides functionality for the user to input audio; and   the search engine:
 identifies compositions that are similar to the audio input by the user; 
 sorts the identified compositions by similarity; and 
 returns the identified compositions to the user via the graphical user interface. 
   
     
     
         26 . The system of  claim 21 , wherein the search engine is further configured to identify similar compositions by comparing the one or more instruments and the one or more ranges of the compositions. 
     
     
         27 . The system of  claim 26 , wherein the search engine is further configured to provide functionality to identify recommendations for the user by identifying compositions similar to at least one composition purchased by the user. 
     
     
         28 . The system of  claim 26 , wherein the search engine is further configured to generate a concert program for the user by identifying compositions similar to one or more compositions identified by the user. 
     
     
         29 . The system of  claim 26 , wherein:
 the system is further configured to store a plurality of user profiles, each of the user profiles including one or more instruments and a range for each of the instruments; and   the search engine is further configured to identify similar user profiles by comparing the one or more instruments and the one or more ranges included in each of the user profiles.   
     
     
         30 . The system of  claim 29 , wherein the search engine is further configured to identify recommendations for the user by identifying compositions purchased by a user with a similar user profile. 
     
     
         31 . A method, comprising:
 storing sheet music data comprising compositions;   storing metadata characterizing each composition, including one or more instruments and a range for each of the one or more instruments;   providing a graphical user interface that provides functionality for a user to identify a user profile that includes one or more instruments and a range for each of the one or more instruments;   identifying compositions by comparing the one or more instruments and the one or more ranges included in the user profile with the one or more instruments and the one or more ranges of each of the compositions; and   displaying the identified compositions to the user via the graphical user interface.   
     
     
         32 . The method of  claim 31 , wherein the comparison of the one or more instruments and the one or more ranges included in the user profile with the one or more instruments and the one or more ranges of each of the compositions uses Manhattan (L1) distance or Jaccard similarity. 
     
     
         33 . The method of  claim 31 , wherein compositions are identified that include one or more parts corresponding to the one or more instruments, where each of the one or more parts are within the range of the instrument corresponding to that part. 
     
     
         34 . The method of  claim 31 , wherein the graphical user interface further provides functionality for the user to search the compositions, the method further comprising:
 generating a query based on the user search;   querying the compositions and the metadata characterizing each composition using the query to generate search results; and   returning the search results to the user via the graphical user interface.   
     
     
         35 . The method of  claim 31 , wherein the graphical user interface further provides functionality for the user to input audio, the method further comprising:
 identifying compositions that are most similar to the audio input by the user;   sorting the identified compositions by similarity; and   returning the identified compositions to the user via the graphical user interface.   
     
     
         36 . The method of  claim 31 , further comprising:
 identifying similar compositions by comparing the one or more instruments and the one or more ranges of the compositions.   
     
     
         37 . The system of  claim 36 , further comprising:
 identify recommendations for the user by identifying compositions similar to at least one composition purchased by the user.   
     
     
         38 . The system of  claim 36 , further comprising:
 providing functionality to generate a concert program by identifying compositions similar to one or more compositions identified by the user.   
     
     
         39 . The system of  claim 36 , further comprising:
 storing a plurality of user profiles, each of the user profiles including one or more instruments and a range for each of the instruments; and   identifying similar user profiles by comparing the one or more instruments and the one or more ranges included in each of the user profiles.   
     
     
         40 . The method of  claim 39 , further comprising:
 identifying recommendations for the user by identifying compositions purchased by a user with a similar user profile.   
     
     
         41 .- 60 . (canceled)