I started to analyze my Google Hangouts history in some of my previous classes. I thought that analyzing conversation history could be a pseudo-natural event that could be turned into a fortune. Although we are always in control of what we type, I believe that when we instant message with those we are closest to, our fingers move without thought.
By analyzing the frequency of messages between myself and two of my ex-girlfriends, I was able to create these songs that represent the unbalance in our relationships.
For every week of texts, I counted the number of characters that I sent and the number of characters that she sent. If I sent more characters, than that week would be represented by a single random guitar note. If she sent more characters, that week would be represented by a single random piano note.
pseudo-code tl;dr:
every note = 1 week of texts
me = 🎸, her = 🎹
if me > her, play 🎸, else play 🎹.
Here is the song that is generated with the messages between me and my most recent ex-girlfriend, GC.
Here is the song generated with my ex before GC.
If you couldn’t tell, the breakup between JJ and I affected me a lot more than that of GC. Both of these songs kind of predict why these relationships ended due to the unbalanced nature of the conversations had.
I decided to sonify the data because I think being able to listen to a whole relationship condensed into a single elegy was more melancholic and romantic.
The majority of the code written lied within parsing the CSV document containing all of the message between myself and the girlfriend. This can be seen here:
https://gist.github.com/bryanjhsu/cfbfdddd073e695acc5683a828154b96