Gesture

Gesture

Overview

Gesture, as seen in the context of computer music composition, is an area of interest actively explored in this wiki. Projects like gest and gestvm work with the notion of Gesture, while other ideas do so more implicitely.

Gesture here refers to a set of specific methods and techniques. The goal of Gesture is to provide simple and intuitive ways to build expressive computer generated musical performances. In particular, there is an emphasis at exploring procedurally generated musical lyricism in synthesized sound (think Clara Rockmore performing "Vocalise" on the Theremin). Use with singing synthesis in particular is a focus, as seen in looptober_2021 and CATFMAS.

General Definition

Websters 1913 Dictionary defines Gesture as "Any action or posture intended to express an idea or passion". In Music, Gesture as a concept is frequently used by performers to help shape phrasing, both in the sound, as well physically on a musical instrument. Gesture implies a continuous motion underneath discrete notes on a page. It is used to give notes context in the overall musical passage. Musicians can manipulate notes with gesture several ways including timing, timbre, dynamics, articulation, and sometimes pitch.

A more common definition is "A motion of the limbs or body made to express or help express thought or to emphasize speech."

Gesture in Computer Music

Gesture in Computer Music tends to map to sounds more directly, rather than being something that is "underneath" a note (several computer music works and composers philosophically unravel the conventional notion of a "note".)

In modular synthesis, anything that is considered a control signal generator has the potential to be used to explicitly create a gesture. An LFO modulating pitch or an envelope controlling filter cutoff could be thought of as a kind of gesture. Just as these components can build up a synthesizer instrument, gestures can coordinate and work together to produce a sound a musical phrase of notes.

This linen of thinking of Gesture as an explicit signal to be used to control parameters is where algorithms and techniques begin to emerge.

Gesture Synthesis and Sequencing

The terms gesture synthesis and gesture sequencing are commonly used interchangeably because of how closely they are related.

Gesture Synthesis refers to the class of DSP algorithms utilized to produce the signal, which should be computed at audio-rate. Gesture synthesizers are glorified breakpoint line generators that use relative timings based on an external clock signal.

The external clock signal, known as a conductor, is a phasor, or a periodic rising ramp signal going from 0 to 1. Phasor clocks can be divided in a generalized way using phase-locked looping. No knowledge of tempo is required, and the process is nearly instantaneous, usually only requiring 1-4 samples of delay for analysis.

Ramps produced from a Conductor are used to interpolate between two discrete values to produce a line. Before interpolating, the ramp can go through a transformation to change the curve somehow, in what is known as a behavior.

Gesture Sequencing the means for telling what the Gesture Synthesizer will perform. This involves programming the breakpoints, durations, and behaviors. In gest, this is done by using commands to produce a tree structure representing rhythmic subdivisions, and targets being at the leaf nodes of the tree. In gestvm, the underlying synthesizer is manipulated using a modified version of the Uxn VM.

TODO Gesture Related Concepts

Using Gesture as a model for structuring music, a few interesting concepts emerge related to timing. This includes Temporal Skewing and temporal weight.

More words to come later.

TODO Pros and Cons

What happens when you think in terms of Gesture? The good, the bad. More words to come later.


home | index