Final month of the project

Entering the final month of this Winter, we also begin the final few weeks of Dispersion.


This project is about exploring possibilities, and enabling a dialogue, both between us the instigators as well as between the sound of the voice and the sound of the bells. We hope to extend this dialogue in the future by including others who are interested in exploring the resonification of the vocal tract, as imagined and articulated through the bells.


The first stage of this would be to make the software available for others to use. This will occur, hopefully later this month when a more functional "user-friendly" version can be built and made available on this site for download.

Terry, 2 August

The lockdown in Melbourne affects the project outcomes

Yesterday the Premier of Victoria declared a State of Emergency, with even further severe restrictions on activities. We were hoping to at least have a workshop on-site, and document it as a final activity for the project. This can now no longer occur.

For the entire duration of the project Carolyn and I have only been in contact by phone, email and the occasional zoom meeting, and we were hoping that by this time the lockdown would have been eased. Instead the opposite has happened. It has been an interesting experience. While it has definitely slowed the creative process down significantly, it has also imposed a certain discipline on our activities. This blog is part of that discipline, and has enabled us to articulate our ideas in perhaps a way that's more formal and focussed than what might otherwise occur.

Terry, 3 August

Repetition variations

Repetition variations with bells octave
00:00 / 02:46

The recurrence of bell strikes can be altered. In this example it ranges from between once every 3 or 4 seconds, to ten strikes per second. 

Vocally, I'm alternating between sustained notes and multiple articulations. Melodically, I'm following the bells pitches, attempting to step away from diatonic and experimental sounds. The result is mostly successful, with sliding pitches helping the unique scale to sound convincing. 


The piece was made by improvising with these guidelines while manually changing the bell repetition dial throughout.

Carolyn, 3 August

Current state of the software

dispersion V0.2 screen shot.jpg

Here's a screen shot of the main windows in the current iteration "Dispersion V0.2". It is starting to mature nicely, but requires simplification, particularly for practical on-site performance applications.


The new module "Voice Straker" sounds great with my test files that Carolyn sent. Can't wait to hear her do some more studies once she has become familiar with the interface. It will be great for more timbral approaches such as ululation, fry and other extended techniques. It also responds well to the spoken voice.

The audio input-output controls are mostly in one window now, and the new tuning meter with the tolerance indicator is incorporated.


Some things to consider for the future:

Vocal gesture recognition and control of the bells through such gestures. I haven't as yet incorporated that into the software. I was hoping to use some machine-learning techniques to achieve this. It will probably not happen unless the project can be extended further at a later date. This approach would then allow the higher-level ideas that were visited in May, like "throwing" the voice, and use of the "tonality diamond" (to systematize the scales available in the bells' tuning) to be incorporated in a concrete, meaningful way.

Integration with a physical interface. I bought a Novation Launchpad with this in mind, so the large number of quite complex parameters in the software can be modified on the fly in a performance setting. It would also be a flexible interface for representing the tuning system of the bells, as a two dimensional grid (like Harry Partch's tonality diamond) instead of as a one-dimensional number-line (like a standard keyboard). However this will have to wait until the project can be extended.

Have the bells emulator record interactions as MIDI files as well as sound files, so the software can be used as a compositional tool, for "through-composed" pieces, with or without a vocal setting.

It would be interesting to apply other sound sources to the software (such as musical instruments). Cello, just-tuned guitar, trombone, even percussion would all be good candidates for future exploration.

At the moment, the triggering of the polytonal bells (bells 8-14) have not been incorporated into the software. This is mainly because the software needs to have another layer of sophistication to accommodate them. The pitches they articulate are included in other bells as individual pitches, but their unique sound is, for now, absent. They would probably be best used in the context of gestural control, or with a particular "tonality" or perhaps used within some sort of rhythmic function, something like a drum kit.

Terry, 3 August

Straker and clicks

Straker and clicksCarolyn Connors
00:00 / 02:47

Straker and clicks uses the Straker application, which is able to produce multiple bell sounds simultaneously, rather than consecutively. My input is all mouth sounds, with no vocal input. The outcome is lots of pretty tinkly, with many beautiful sustained bell chords.

Carolyn, 15 August

Version 0.3 of the software done

So I've cleaned up the interface, and simplified it somewhat. Hopefully it will be a little more intuitive to use. There were plenty of bugs that I discovered and fixed along the way. You should be able to experiment with the Voice Straker module, and are now able to save settings that you like. Also the polytonal bells (bells 8-14) are now incorporated into the Straker's output (but not the Pitch Tracker or the Harmonizer). This gives a richer sound for the Straker. Here's a demo video.

Terry, 15 August