Audio Reactive Processes

Hello, I am a Master of Architecture student currently completing my thesis on audio as a tool to alter spaces in real-time, kinetic architecture with a focus on audio reactivity.

I have recently picked up vvvv gamma and have been using to attempt to create audio reactive 2d and 3d forms. I have been using the audio player as input then the audio is sent through the FFT node which I then average the values to give a single float value. The problem with this is that the value change is quite sharp; The values change too quickly, so when I assign the averaged float value to any form the change is too sharp and janky.

Does anyone know how to smoothen the value change, or perhaps have a suggestion for a completely different approach to audio reactivity?

Hey, welcome! You can use a Damper to smoothen the values over time. Give it just a short period (like 0.2 seconds) and the value will be much smoother.

1 Like

and a reminder, since you’re new: open the helpbrowser (pressing F1), switch to the “Learn” tab and type what you’re looking for. in your case typing “smooth” would have lead you to a help patch “HowTo Smoothen a noisy signal”.

1 Like

Thank you both, the damper worked quite well, and I will make the most of the “Learn” tab.

1 Like

You are on the right path and FFT is one way to get something audio-reactive. It’s always the question how much “data” is in FFT and that depends very much on the source material. The more instruments, samples, voices, etc. you have in the audio source, the less meaningful the data that FFT can represent will be.

If you are interested, we have been working on “altering spaces in real-time” coming from an architecture background for the last 20 years or so and for the last 10 years using vvvv.

You can find one of our old TEDx talks about it here: Musical Sculptures - Jan Henrik Hansen at TEDxZurich It used to be on Youtube, but now isn’t there any more, so I re-uploaded it to our Vimeo. You can see me during the live presentation using the old vvvv beta and midi controllers.

Nowadays SpaceMusic is much further along, infinitely bigger and more capable and getting very close to a 1.0 “release”. I say that in quotes as it will probably not be released as a software as such. We recently used it for a project for ABB in Switzerland, where it is used for a permanent installation running on an LED screen and translating the buildings automation data.

You can see it in action a little bit more here: Videos Sorry for the lack of information on it, the website is still very much work in progress and only has a few videos on it.

1 Like

There is also the SpaceMusic presentation from Node20:


@seltzdesign Thank you, I am looking at using ambient sounds collected from natural and urban environments. I am using 4-channel audio. Your program, SpaceMusic produces some stunning results!! Would there be any possibility of accessing this tool? Thank you for your response!

@bjoern Great find, thank you!

For now it is still work in progress unfortunately. There will also be a new presentation at some point soon. It will hopefully become available as a closed alpha in some way. I will let you know personally about any developments.