Thoughts on Sound Diffusion V

Listening to a multidimensional representation of sound won't be as accurate or complete as what our ears perceive if it were a 'real' sound source which will present an issue in creating sound spatialisation. So for me it’s a twofold consideration: of what and how that sound is placed in space equally as what implications of space are in the sound source itself.
Knowing a piece well enough to do justice as the composer intended is a challenge in sound diffusion nevertheless our interpretations in sonic art group have been interesting. The next step is to perhaps create a musique concrete piece ourselves taking sound spatialisation into consideration in terms of the creative process.
I hadn’t considered using a number of different spatial formats for an electroacoustic piece. A six channel stereo system controlled within a multi dimensional interpolation space along with fixed sounds of mono tracks created in Nodal and a separate control for other mono tracks would give greater range in representing what we wanted musically.
What captures my imagination are the creative possibilities in working this way. For example, the use of a wiimote (I believe the new iphones also have an accelerometer in them) in conjunction with other sources and layers of sound would give the composer an opportunity to use it as a live performance tool.
Obviously there are technical hurdles to overcome in implementing technology such as this eg. ability to pick up an accurate and consistent signal and the issue of latency, but not only could it enable the audience to visualise the directionality of a sound source controlled and directed by the composer/performer using a remote device but also allow for greater musical expressivenesss and flexibility in a performance.
Without taking away from notion of a musical piece being valid on its own merit, a visual perspective may enhance an auditory experience which, in turn, could make cross media pieces such as this more memorable and meaningful.

Thoughts on Sound Diffusion IV

My interest in sound diffusion is in its use as a sound parameter for compositional manipulation. Traditionally, the primary parameter[1] has been pitch, with volume and timber close behind. Over time, a rich system of language has developed around the use of pitch in music. Modes, scales and harmony are our method of creating organised, ineligible sound from the continuous choices in pitch from (almost) 0 Hz to 20kHz. Likewise, volume is organised into categories such as pianissimo, mezzo forte and sforzando[2], and as composers we (try to) understand how to use volume to support a composition.

As a compositional parameter, though, diffusion is probably not as well understood. Avoiding the temptation to grandstand about the opportunity to create new knowledge, I think this relatively unexplored parameter provides a kind of freedom to discover a language without the legacy of an established body of knowledge. So far we tend to use diffusion somewhat crudely - mostly to support a sense of depth (more distant sounds are made to be more diffuse). Of course, this is due to the positive correlation in the natural acoustic world between diffusion and distance.

Must we be limited to this though? What if we manipulate diffusion to support the dynamics of a composition itself? In an auditorium filled with speakers (and listeners!), the temptation to correlate the number of speakers to aural size may be difficult to resist, but what about an inverse correlation? For example, one might establish a sonic language of quiet, spatially diffuse sound (perhaps using grain clouds sourced from an outdoor recording). Following this with a direct sound (such as a sawtooth wave at 1kHz), whilst simultaneously collapsing the spatial field down to a single speaker could produce quite an arresting effect. Of course, there are many more possibilities worth exploring.

-Kim.

[1] I'm not counting time as a parameter for the purpose of this discussion, as it behaves differently to these parameters of sound.

[2] Or more recently, K-20/RMS, K-14/RMS and K-12/RMS.

Thoughts on Sound Diffusion III

The stereo space can replicate or imitate a myriad of environments. With two speakers a sound can be given depth, size, position etc. using level, panning and reverb. Using a live multi channel mix in a performance area as a way of shifting the environment of a stereo sound takes these aspects further. Depth cannot only be implied by manipulating level and reverb, but can actually be achieved by changing the position of the sound course in relation to the audience.

The sounding effect may be achieved in the production of recorded sound but the very fact of the source of the sound changing in front of an audience using different speakers constitutes a radically different aesthetic, as well as illuminating the aesthetic of acousmatism to greater effect.

This difference in aesthetic is for me what is interesting about sound diffusion, rather than the ability to make sounds whir around a space, or to achieve a multichannel sounding piece using only two channels of material.

To achieve my preferred effect created through sound diffusion it is essential that in addition to differences in positioning of the different speakers, other elements must also be at work. For example, it is useful for the performance space to lend itself to different kinds of reverb, and for the placement of speakers to exploit this; the speakers themselves can be of different sizes, set to different levels, of different bandwidth, so as to be able to maximally demonstrate different moods and so on in from the sounds to the space.

Acousmatic Art:
http://www.personal.leeds.ac.uk/~muswlw/pubs/lwthesis.html

Thoughts on Sound Diffusion II

When we first started doing sound diffusion, I thought that all we would really be doing is expanding a normal stereo speaker system from 2 to 6 speakers. I was not all that sure about what we would actually accomplish. However once we set up the system, and introduced the MIDI sliders to control the levels of each speaker, I knew that there was potential.

What set sound diffusion apart from a multi-channel piece of music for me was the live performance aspect it allowed me to explore. It gave me the opportunity to take an existing recording of a piece of music, and re-perform it by controlling the dynamics, and spatial placement of the sound. I could make a sound start from the back of the room, and slowly (or rapidly) bring it out, sending the sound skyward, expand it out to the sides, or both.

Until now, I had only really worked with how you could emulate space in music by changing parameters such as volume, reverb and panning. At the beginning of this project, I did not think that having multiple sound sources would be very different to what I had done. However, I quickly found that I was wrong, and that there were many aspects of physically placing a sound in a space that I could not emulate. The room contributes to the performance, as you can hear the sound bouncing off the walls, the ceiling and the floor, something that would be extremely difficult to emulate, and almost impossible to do realistically. The freedom to move the sound anywhere whenever I wanted truly made me feel as though I was in control.

Thoughts on Sound Diffusion I

I think that one of the nice things about sound diffusion that struck me when the concept was first introduced is the conceptual simplicity of it. We spend a lot of time and thought within the world of music technology trying to invent completely new ways of doing things – be they innovative instrument or synthesis design, new methods of signal processing, alternative controllers and cross-media practices. As a result it is refreshing to use an approach that is really not much different from the principle of multi-track recording – albeit in a ‘reverse-engineered’ manner – to manipulate and spatialise sound.

Taking stereo audio channels and spreading them around a variety of speakers in an auditorium also emphasises the importance of acoustic space. Having memorised a track in its normal stereo configuration and getting to know its intricacies this way, and then placing it in all corners of a hall can greatly transform the work. You start to hear the track in completely new ways – indeed it can become a new piece altogether. In a way, it can be something like a process of live re-composition, or at least remixing of a piece.

This also provides a simple way in which you can ‘play the room’ in which you are listening to the piece. The room’s acoustic becomes a significant factor in the process of diffusion. How this is manipulated and played upon by the diffuser is part of the enjoyment of the experience. It also means that the role of the audio engineer in the production of the recorded music is rendered slightly less important, as the previously definitive recording of a work is now open to reinterpretation.

The variable acoustic responses of different sets of speakers is also striking when doing a diffusion. This is a feature that can be exploited by the diffuser to creative effect. In particular, in our experience, the upper speaker pair in our six channel array is of significantly lower acoustic quality than the two Bose pairs. Their frequency response tends to be significantly in favour of the higher end of the frequency spectrum. This has an effect of emphasising the closeness of the sound – which actually provides a great contrast with the boomier characteristic of the auditorium. This contrast, particularly when listening to ‘smaller’, less reverberant sounds can be used to great effect.

Of course, not every piece of music ever written is ideal for sound diffusion. This much has become apparent in our first few experiments. Obviously, the technique has a natural affinity to the musique concrete end of the musical spectrum. I think the reason for this is that this genre places the emphasis on the quality of individual sounds, and often exploits a wide variety of different sounds. The other significant factor in musique concrete is there is often quite a deal of space (silence) within the music. Musical space (not spatialised space) is really helpful when doing a diffusion, as it gives you a chance to rapidly shift the position in acoustic space. This allows a point of tension to be created as it becomes uncertain where new sounds will be heard. Also, sounds the listener is accustomed to might begin to appear in a completely new context. Having said this, it is also a good effect to ‘ride the levels’ of the different channels for a continuous sound within the musical texture and bring it in and out of various spatial configurations.

sound Diffusion Intro

After attending a workshop held by Robert Normendeau at the this years ACMA conference I became aware that sound diffusion is a useful approach to exploring electroacoustic music in a learning environment. So this semester the Sonic Art Group started investigating sound diffusion. I made a Max patch that mixes stereo to six channels. The levels are controlled by MIDI sliders and the system works reasonably well. In the image above is a sonogram of Bernard Parmegiani's wonderful acousmatic work Incidences/Résonances which lends itself very well to diffusion and we started with this piece as well as Tristram Cary's Continuum (a synthesis piece from 1968 which is still remarkable fresh and imaginative). Following are reflections by those who have been involved. You can sample Parmegiani's work at last.fm, see:

http://www.last.fm/music/Bernard+Parmegiani

So what is sound diffusion? According to Barry Truax its:

Electroacoustic diffusion refers to the practice of distributing sound throughout a space using multiple loudspeakers, usually with real-time control over the sound levels, equalisation, and placement of the sound. The source material may be live performers, synthesizers, and/or stereo or multi-channel tape recordings. (Source: Barry Truax - Handbook for Acoustic Ecology CD-ROM Edition. Cambridge Street Publishing, 1999 - CSR-CDR 9901)

Usually an existing stereo recording is "diffused" out to a multispeaker array. Instead of mixing multiple channels of sound into two, here you take a two channel stereo format and mix out to multiple channels. At first glance this approach seems a little redundant but what becomes apparent is that the variable size of the sound through speaker array and its presence (via resonance) in the acoustic space can be powerful ways to render a fixed recording. In fact the live diffused performance is actually a interpretation of the work being played. From a teaching and learning perspective this is very interesting because the discussion that we have is focused (in the main) on musical issues instead of technical ones.