AI Summary of Scholarly Research
This page presents an AI-generated summary of a published research paper. The original authors did not write or review this article. See full disclosure ↓
Publication Signals show what we were able to verify about where this research was published.MODERATECore publication signals for this source were verified. Publication Signals reflect the source’s verifiable credentials, not the quality of the research.
- ✔ Published in indexed journal
- ✔ No retraction or integrity flags
Overview
String Theory is an interactive sonification installation that transforms naturalistic conversation data into real-time audio performance through participant engagement. The system employs the CANDOR corpus, comprising over 1,500 unscripted conversations between strangers, as its foundational dataset. Participants interact with the installation via networked mobile devices connected to a loudspeaker orchestra configuration. The system applies parameter mapping sonification techniques to organize and synthesize participant interactions, generating spontaneous collaborative audio experiences that reflect interpersonal dynamics.
Methods and approach
The installation architecture integrates a loudspeaker orchestra setup with network-based mobile device connectivity to enable real-time participant interaction. The CANDOR corpus provides empirical grounding in naturalistic conversational patterns between unfamiliar interlocutors. Parameter mapping sonification methodology translates conversational and interactive features into acoustic parameters, with algorithmic processes determining arrangement and synthesis of audience contributions. The system operates through computer-mediated synthesis that responds to networked input in real time, producing unique performances that vary based on participant behavior and interaction patterns. Technical implementation details and setup protocols are documented in an open-source GitHub repository.
Key Findings
String Theory demonstrates a functional sonification system capable of rendering conversational dynamics through acoustic performance in real time. The installation framework enables audiences to collectively participate in and perceive interpersonal interaction patterns through synthesized sound, creating emergent acoustic structures that reflect both individual and group-level behavioral contributions. The system's design accommodates variability across performances while maintaining consistent mapping between conversational phenomena and sonic output.
Implications
The work establishes sonification as a viable method for revealing structural patterns within naturalistic conversation data, particularly regarding dynamics of unfamiliarity and vulnerability in interpersonal exchange. By translating conversational datasets into participatory acoustic experiences, the installation bridges empirical research on human communication with artistic and experiential inquiry, enabling audiences to encounter conversation dynamics through non-visual sensory modalities. This approach extends sonification methodology beyond exploratory data analysis toward interactive knowledge production. The open-source availability of the system facilitates extension and adaptation across institutional and artistic contexts, supporting broader investigation of sonification applications in studying conversational phenomena.
Scope and limitations
This summary is based on the study abstract and available metadata. It does not include a full analysis of the complete paper, supplementary materials, or underlying datasets unless explicitly stated. Findings should be interpreted in the context of the original publication.
Disclosure
- Research title: String Theory: Participatory Sonification Performance of Naturalistic Conversation
- Authors: Tristan Peng
- Institutions: Georgia Institute of Technology
- Publication date: 2026-03-07
- DOI: https://doi.org/10.1145/3731459.3779146
- OpenAlex record: View
- Image credit: Photo by Michael Lee on Unsplash (Source • License)
- Disclosure: This post was generated by Claude (Anthropic). The original authors did not write or review this post.
Get the weekly research newsletter
Stay current with peer-reviewed research without reading academic papers — one filtered digest, every Friday.


