Abstract
Characterizing the acoustic environment of a geospatial region may be challenging due to limitations in some existing data sources, which can be outdated, geographically broad, subjective, or anecdotal. This document describes a multi-agent computational framework for dynamic geospatial acoustic characterization. The system can utilize a collection of specialized software agents to ingest, process, and synthesize data from multiple heterogeneous sources, such as street-level and aerial audio recordings, user-generated content, and official data feeds. For example, agents can perform tasks such as data sourcing, denoising, source-specific characterization (e.g., vehicular, industrial, human), and temporal synthesis. This framework can be used to produce a detailed and temporally-variant acoustic summary for a given location, which may result in a characterization with increased granularity and contextual richness compared to methods that could rely on singular or less diverse data sets.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Melvix, Lenord, "A Multi-Agent Framework for Dynamic Geospatial Acoustic Characterization", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/9204