I play music for some of our church services, and lately I’ve been playing for about 75% of them. My normal process is to look at the scheduled hymns and then look through all of my piano music to see if I have any arrangements of songs that match the hymns so I can use them for pre and post service music. Every time I buy a new piano book, I put it into a spreadsheet that matches the sheet music tunes to the hymn tunes so I can look them up easily in the future.
Now that I’ve been playing more Sundays, I’ve been looking to fill in some holes in my collection. But which piano book would fill the most holes? Sounds like a good programming problem.
I started by feeding AI the shared spreadsheet that the church worship team uses. I had it analyze all the hymns we’ve used for the last three years and then figure out how often we use them. Then I fed it my spreadsheet of all the music I already have. Then I pointed it to some online shopping sights from my favorite arrangers.
Using all that information, it assigned a point value to each hymn based on how often we’ve sung it in the past year and then filtered out any hymns that I already have covered with some other sheet music. Then it looked through the online catalogs of all the arrangers I like and, in the end, it spit out a list of books I should consider buying and told me who many “points” they would cover in my library.
I ordered the books a few minutes after I saw that report and later it dawned on me that maybe I should have validated the results before spending money based on what AI just told me. But it worked out great! The new books were excellent ways of giving me pre and post service arrangements that match some of our most-used hymns.
If you also play piano and are curious what arrangers I pointed the AI to, here is a list:
There are a lot more that I like but that seemed like a good enough sample size to prove whether this worked and I will probably expand it more in the future if I try to execute this again.