for one or more performers
a sound score for listening, memory, and response
Rob Canning
What You Heard is a quiet, open-form composition for one or more performers, realised as a sound score built from textual prompts, sonic memory, and replayed recordings. The work does not prescribe musical material or a fixed temporal form. Instead, it establishes conditions under which performers listen, recall, respond, and allow sound to circulate as memory rather than as event.
The score functions as a single mutable palette of prompts and traces. Performers do not move through material in a linear order but establish chains of events through listening and response. Form emerges through accumulation, hesitation, interruption, and restraint rather than through development or narrative progression.
The work may be performed solo or by multiple performers simultaneously. In all configurations, listening precedes action, and silence is treated as an active musical state.
The primary notational material of What You Heard consists of short textual prompts positioned freely on a scroll- or page-based score surface.
Some prompts appear descriptive, others evocative or situational; onomatopoetic cues occupy a middle ground in which language already sounds. In practice, these distinctions collapse. Even the most specific descriptions rely on memory and prediction, while evocative cues imply particular sonic contours.
The performer does not execute instructions so much as negotiate them through listening, recollection, imagination, and attention to the present acoustic situation. Text functions not as representation but as a trigger for auditory memory. Reading becomes a form of listening in advance; performance becomes an act of remembering rather than reproducing.
The work explores interstitial spaces in performer–composer systems of communication, where sound is neither fully specified nor entirely free. Meaning is distributed across text, sound, memory, and situation rather than encoded symbolically.
The score does not distinguish sharply between sounded and unsounded material. Instead, it presents a continuum of states:
- suggested
- predicted
- enacted
- recorded
- replayed
- remembered
These states may coexist or overlap. Uncertainty is treated as a structural feature rather than a problem to be resolved. Listening precedes action, and hesitation, restraint, and silence are integral musical materials.
The initial set of textual prompts provided by the composer is not fixed. Performers may subtly or radically alter the aesthetic of the work by adding their own prompts—single words, phrases, situations, or memories—directly into the score.
Through this process, performers inscribe their own memory language into the work. The composition provides a framework for eliciting sonic memory rather than a closed inventory of material. Different realisations may therefore diverge significantly in tone, density, and affect, even while sharing the same underlying structure.
As an optional extension of this openness, performers may retitle the piece using one of their own prompts. What You Heard names the work at the point of composition; a realised version might instead be titled, for example, What I Heard When I Lost All Hope or What You Heard in the Next Room. In this way, even the identity of the work may emerge from the material it generates.
In performance, performers respond to textual prompts by producing sounds—instrumental, vocal, or environmental. These sounds may be ephemeral or may be recorded and reintroduced into the score.
Replayed sounds function as sonic memory traces rather than fixed musical material. They may be layered, revisited, degraded through repeated playback, or allowed to fade into silence. Over time, the piece forms an evolving archive shaped by repetition, loss, and selective recall rather than linear development.
Sound thus operates both as action and as memory: something heard, something half-remembered, something no longer certain.
What You Heard is conceived not only as a single performance but as a work that can unfold across iterations. The score may include explicit cues such as:
performer replays previous iteration and listens
performer rewinds, replays previous iteration, and responds
Through these actions, material from earlier rehearsals or performances can shape later ones. The work becomes diachronic: each iteration leaves traces for the next. Memory is not preserved as an archive but reintroduced as a mediated, imperfect presence.
The work deliberately accommodates simple, external recording devices—such as mobile phones or dictaphones—as valid components of the piece. These devices function as informal memory tools, allowing performers to capture sound quickly without needing to engage deeply with the score system.
This lo-fi mediation offsets the use of digital score technology and reinforces the work’s concern with memory as a situated, embodied, and imperfect process. Fidelity is not a goal; immediacy and attentiveness are.
Some realisations of the work distinguish between externally projected sound and privately heard material. Certain cues or replayed sounds may be routed at very low level to a single in-ear channel, while others are diffused through the performance space.
This separation reinforces the work’s concern with private versus shared listening, and with sounds that are remembered internally rather than overtly present. The technical realisation is intentionally simple: a passive stereo splitter may route one channel to a small loudspeaker and the other to a single earbud. No specialised monitoring system is required.
Although conceived for solo performance, What You Heard may also be realised by multiple performers simultaneously. In this configuration, the work becomes an ensemble piece in which sonic memory is distributed across several listening positions rather than synchronised into a single shared part.
Each performer works with their own instance of the score, private listening channel, and small loudspeaker. Textual cues authored or adopted by different performers are visually distinguished by colour, allowing multiple memory streams to coexist on the score surface. Colour functions not as role assignment but as a marker of authorship and perspective.
Performers may trigger their own cues or those introduced by others. A single cue may be directed either to shared sound or to private listening. In some realisations, cues are spatially divided so that interaction on one side of the text produces externally diffused sound, while interaction on the other produces sound audible only to the triggering performer.
Through these mechanisms, ensemble performance explores partial knowledge, misalignment, and asynchronous listening. Performers inhabit overlapping but non-identical sonic worlds, attending not only to what is sounded in the room but to what remains private, delayed, or unheard. Coordination emerges through listening and restraint rather than through synchronisation.
What You Heard does not unfold along a predetermined timeline. Instead, the score remains as a single mutable palette of prompts and sonic traces from which performers establish chains of events through listening and response.
No fixed durations, sections, or ordering are prescribed. Events may trigger further events, be replayed, be withheld, or be allowed to dissolve into silence. Silence is not a transition but an active state within the chain.
Form emerges through accumulation, interruption, and restraint rather than development. Private listening may precede public sound; replay may replace new production. The piece remains open to stasis as much as to change.
Although no fixed rules are imposed, performers may adopt simple listening-based strategies to shape chains of events. These strategies function as shared values rather than instructions, for example:
- allow each sound to decay fully before responding,
- replay an existing sound before introducing a new one,
- let private listening precede public sound,
- allow silence to interrupt or dissolve any chain.
The absence of response is treated as a valid action within the work.
Only the conditions for beginning and ending are fixed.
The work begins by entering a shared listening state, typically in silence or near-silence. Performers orient themselves to the space, to each other, and to the available prompts before producing sound.
The work ends when performers collectively recognise that no further response is necessary. Sound is allowed to withdraw without cadence or closure. Ending is understood as a shared decision rather than a timed conclusion.
While the work avoids a fixed timeline, it may include lightweight mechanisms for shared orientation. A common realisation uses a single visual contour traversed slowly from beginning to end, functioning as an attentional guide rather than a sequence of musical events.
In this configuration, a simple object moves along a curved path beneath the field of textual prompts. The horizontal axis loosely corresponds to elapsed shared time, while the vertical dimension suggests degrees of activity or density. When the contour is high, overlapping sound and collective activity are permitted; when it is low, restraint, listening, and space are prioritised. The contour conditions behaviour without prescribing it.
Along this path, one or two structural markers may be placed. These indicate shared actions rather than musical content, for example:
- all performers begin recording for a fixed duration,
- one performer stands and occupies the foreground while others withdraw,
- all performers pause and listen without producing sound.
These actions function as coordination points and choreographic gestures rather than cadenzas or formal divisions.
The score is realised as an importable Oscilla project file (what-you-heard.oscilla) accompanied by freely positioned textual annotations on a scroll- or page-based surface. No traditional staff notation is used.
Oscilla runs locally in a web browser and supports executable cues, shared timing, and audio playback. A shared, network-synchronised stopwatch may be used for rehearsal and orientation without prescribing musical form. An internet connection is not required during performance.
All materials—score, documentation, and example media—are hosted at:
https://github.com/robcanning/what-you-heard
- The piece prioritises listening over projection.
- Sounds should be quiet, fragile, and responsive to the acoustic environment.
- Instrumentation is open: acoustic instrument, voice, objects, or environmental sound are all appropriate.
- Performers are encouraged to add their own textual prompts.
- Interaction with technology may be minimal; simple recording devices are welcome.
- Replay is optional; silence and restraint are integral to the work.
Typical duration: 10–20 minutes, flexible.
Provided by performer(s):
- Tablet or laptop (tablet preferred)
- Modern web browser with Web Audio support
- Oscilla software (standalone binary or local server)
- Mobile phone or portable recorder (optional)
- Simple earbud (single ear)
Required at venue:
- Small PA system suitable for low-level sound
- Power supply near performance position
- Quiet performance environment
Preferred but not required:
- Passive stereo Y-splitter from device headphone output
- One channel routed to a small loudspeaker
- One channel routed to a single in-ear earbud
- Local Wi-Fi router (LAN only; no internet required)
No specialist technical assistance is required.
What You Heard forms part of a broader research-led artistic project exploring executable sound scores, sonic memory, and listening-led performance practices. Related submissions include:
- Paper: Audio as Notation: Embedding Sonic Processes in Oscilla’s Graphic Score Environment
- Workshop: Executable Sound Scores with Oscilla
The compositional framework is complete. The work remains open to ongoing realisation through performance practice, iteration, and performer contribution.
Rob Canning is a composer and artist working with graphic notation, sound scores, and networked performance systems. His work explores listening-led notation, executable scores, and collaborative performance practices across acoustic and electronic media. He is the creator of Oscilla, a browser-based environment for animated and sound-driven graphic scores. His work has been presented in concert, installation, and research contexts internationally.