What I learned
Through this project, I learned so much about so many different things.
- I learned how to use Garageband, including how to allow participants to hear both themselves and a recording while they sing. In this class we learned how to use lots of different types of online platforms for working with audio signals, so I used some of that technical knowledge as I tried to figure out how to use Garageband.
- I learned how to write Praat scripts. This was very helpful because I didn’t need to manually pull the formants out of every recording, which I thought I was going to have to do.
- I gained lots of experience coding with Python, since this was my first coding project not for a CS problem set, exam, etc. I applied some of the code that I learned in this class, as well as lots of code I learned this semester in 6.00 at MIT.I learned the importance of organizing data in a functional and easily understandable way. I also learned the importance of using documentation, as well as learning from (while citing!) programs that others have already written.
- I also learned that there may be a correlation between F1 and blend in vocal music, which I didn’t know would be the case. I thought F2 would have a bigger effect because F2 is generally affected by lip routing, which is a big part of blend. I applied what I learned about waves in this course as I considered the fact that each formant is made up of multiple harmonics. For example, F1 does not equal the first harmonic.
- I learned that participants’ formants vary wildly! This is evident from the huge differences in formants in some of the graphs, especially in /i/ in the F2 graph. I wonder why the wave frequencies differ so much on this vowel in this formant.
In future classes
I would have liked to learn a bit more about human vocals in this course. It would have been cool to use Praat in the class to look at our own vocal recordings and to consider how formants have to do with harmonics and waves. I thought it was fascinating to look at waveforms for the JFK conspiracies at the end of the course, and it would have been cool to use waveforms and spectrograms more in the class.