Phonosynthesis: Sound Bouquet
When working, Sound Bouquet performs a process of phonosynthesis in three phases:
1. Sound Bouquet collects sounds from the environment – 5 to 10m away from the sculpture two microphones are placed. They are intented for the audience interaction. Third microphone is placed somewhere in surrounding area so it can gather the ambience sounds.
2. Through predefined MAX/MSP algorithms Sound Bouquet organises sounds into soundscapes – the algorithms that process human sounds (gathered from two audience microphones) use set of rules to build simple rhythmic phrases. These rules can be easily discovered by the audience member playing with the sculpture, alone or with a friend/partner in front of the other microphone. The algorithm in charge of processing the general ambience builds drone-like musical atmospheres, changing moods from frightening to peaceful.
3. Sound Bouquet diffuses musical composition into surrounding space – the sound sculpture is built from 9 horn speakers and 1 subwoofer. It can cover a circular space with radius of 5 to 100 meters, depending on the size of the space, wanted coverage and microphone placement.
This is a great idea. I wish I had the money to do a similar project! I love how well it sounds.
Actually, can I ask you a question?
I'm doing an interactive project at the moment myself and I was wondering how you prevented the sound of the speakers from bleeding into the microphone for further processing. Did you use gates? Amplitude? Or are the speakers positioned in such a manner that it prevents leakage?
Regards,
John.
it has a simple input gate, and when input level crosses a treshold, it triggers a two-second recording buffer, which then loops at various speeds. and yes, i achieved separation by using speakers' and microphones' directivity.
I am doing exactly that too - using a 1.5 second buffer which is retriggered, cleared, and rerecorded over and over again. I will have to look more into speakers’ and microphones’ directivity.
Thank you,
John.