CMU School of Drama


Monday, March 29, 2021

The Wonder Of Wavelets: A Tool To Clarify The Difference Between Delay, Phase Shift & Polarity

ProSoundWeb: Wavelets are both a theoretical tool and a system tuning tool. Here’s how to use them to address some common system tuning problems. In the time domain, we are blind to frequency. In the frequency domain, we are blind to time (Figure 1). This results from one of the most important relationships in audio: TF = 1

2 comments:

Alexa Janoschka said...

I feel smarter after reading this, even though I am still confused about what is going on I enjoyed reading it. The first part was hard to understand, especially not knowing much about sound engineering myself. I found the Takeaway section and the Conclusion section very helpful when trying to wrap my head around the information a little more. The graphics were also very helpful while trying to understand the theoretical concepts that the author was talking about in the article. I have heard of the word phase shift used in a lighting context but I don't know how different it is within the world of sound. These articles make my brain hurt but at the same time remind me that there is a lot that I can learn about which is excited but also stressful, haha. I liked the author's addition of the feeling of discovering something new or finding out that you have have cool tools but never knew how to use them, that's always a fun feeling :)

Elliot Queale said...

This was a surprisingly awesome article that actually has so much application outside of pure sound design. The opening of the article starts with a discussion about the difference between the time and frequency domain, which is one of the core concepts of classical control systems engineering. It also, oddly enough, is analogous to the Heisenberg uncertainty principal in quantum mechanics (i.e. we can know a wave's frequency, but it doesn't really have a position, and similarly we can find a peak amplitude of a wave, but we have less knowledge about its frequency). ANYWAY, the idea of using these little wave packets (dubbed wavelets) looks like a great solution to the challenges of setting up a system. First, the discrete signal packet acts like a clapboard when setting up speaker delays, which is one of the more obvious applications. The other two, however, allow us to see exactly how the room acoustics are affecting that specific frequency we are sending. I especially like the explanation of the APF vs delay as analogous to hammering a nail vs screwing a screw into a board. Now that I no longer do much when it comes to audio, I don't know if and when I would apply this. Nevertheless, I really think it is a cool topic that does bleed into other aspects of our industry, especially electronics.