Noise pollution is one of the most serious quality-of-life issues in urban environments. In New York City (NYC), for example, more than 80% of complaints1 registered with NYC’s 311 phone line2 are noise complaints. Noise is not just a nuisance to city dwellers as its negative implications go far beyond the issue of quality-of-life; it contributes to cardiovascular disease, cognitive impairment, sleep disturbance, and tinnitus3, while also interfering with learning activities [21]. One of the greatest issues in measuring noise lies in two of the core characteristics of acoustic noise itself — transiency and structural multidimensionality. Common noise measurement practices based on average noise levels are severely inadequate in capturing the essence of noise and sound characteristics in general. Noise changes throughout the day, throughout the week, throughout the month, throughout the year, and changes with respect to its frequency characteristics, energy levels, and the context in which it is heard. This paper outlines a collaborative project that addresses critical components for understanding spatiotemporal acoustics: measuring, streaming, archiving, analyzing, and visualizing urban soundscapes [28] with a focus on noise rendered through a cyber-physical sensor network system built on Citygram [23, 24].

Automatic urban sound classification is a growing area of research with applications in multimedia retrieval and urban informatics. In this paper we identify two main barriers to research in this area - the lack of a common taxonomy and the scarceness of large, real-world, annotated data. To address these issues we present a taxonomy of urban sounds and a new dataset, UrbanSound, containing 27 hours of audio with 18.5 hours of annotated sound event occurrences across 10 sound classes. The challenges presented by the new dataset are studied through a series of experiments using a baseline classification system.

This paper presents an exploration platform for locative sonification based on audio feature vectors extracted from urban spaces. Our locative sonification research is part of a larger project called Citygram[17]. Citygram focuses on geospatial research that is concerend with automatically collecting, visualizing, analyzing, and mapping nonocular energies from urban environments. The acoustic data is captured via off-the-shelf poly-sensory Androidbased remote sensing devices (RSD). Audio feature vectors are streamed to and stored in the Citygram database which can then be used for sonification and visualization. The first iteration, Citygram One, concentrates on urban acoustic energies, rendering spatio-acoustic feature vectors with the aim of better understanding our environment, large cities in particular. This paper focuses on using the Citygram framework for creative practice via locative sonification.