This post is the first in a series of posts on 3D Audio. The next post in the series is Investigating 3D Audio: How We Hear.


Back in October, I had a thought.

Wouldn’t it be cool if I could work 3D audio into my compositions?

I envisioned all sorts of interesting ways I could incorporate 3D audio into my music making. Most involved applying 3D effects to my music through the use of algorithms. What I realized, after lots of daydreaming, was that I do not truly know how we hear in 3D. What sort of signals do I need? How can I trick an ear?

Luckily, the Immersive Music Hackathon was coming up. This gave me sufficient motivation to look up psychoacoustics, figure out what sort of concepts I would want to know to start to tackle a project for the hackathon. As an organizer, I would not have much time to work on the project while at the event. But perhaps I would have enough time to work on a scheme for creating 3D audio appropriate for my music.

While the hackathon was a resounding success, and Alex, Charles, Jason, Paul, Rachel, and Thor were amazing people to bounce ideas off of, I ended up needing a bit more time to do the research. Some concepts I was introduced to at the hackathon are HRTFs and ambisonic technology. While I wasn’t introduced to it at the hackathon, I was reminded of the Doppler effect and it’s uses in audio. Other concepts I will expound upon follow from my investigations into these topics.

Investigating concepts

When investigating concepts, I aim to keep the following questions in mind:

  • Have I learned enough about how, psychologically, we humans process sound to use my knowledge to make interesting artistic effects?
  • Have I found pre-existing solutions that will make implementing my ideas easier?

I routinely use Python 3.6 to analyze data, but could also use the R Programming Language to analyze data. I have used MIT’s music 21 Python library to generate music within python in the past, but as implementing multi-channel solutions is not a feature of music 21’s, I need to generate audio in some other fashion. As I am comfortable with Python, and know that I can use it as an interface to Csound to create music, and the Csound library is well-maintained and has a wealth of functions I can draw upon, I will search for solutions that will be possible using a mix of both programs.

Exploration of concepts will be continued on another day, in Investigating 3D Audio: How We Hear.
Last edited – January 13, 2018.