Environmental Sound Perception: Metadescription and Modeling Based on Independent Primary Studies

Abstract

The aim of the study is to transpose and extend to a set of environmental sounds the notion of sound descriptors usually used for musical sounds. Four separate primary studies dealing with interior car sounds, air-conditioning units, car horns, and closing car doors are considered collectively. The corpus formed by these initial stimuli is submitted to new experimental studies and analyses, both for revealing metacategories and for defining more precisely the limits of each of the resulting categories. In a second step, the new structure is modeled: common and specific dimensions within each category are derived from the initial results and new investigations of audio features are performed. Furthermore, an automatic classifier based on two audio descriptors and a multinomial logistic regression procedure is implemented and validated with the corpus.

Prioritizing foreground selection of natural chirp sounds by tempo and spectral centroid

“…We argue that, beyond loudness, tempo and brightness are the next important dimensions to predict stream selection and the formation of the auditory foreground. These perceptual features are also the most important ones in similarity judgments [28], especially for environmental sounds, which are the focus of our study.…”

“…In the remainder of this section, we suggest that we should distinguish between the salience of events and that of sound streams, i.e., the auditory objects [2] in a scene. Then, starting from a set of features motivated by research on timbre perception [28,29], we select those that we hypothesize to be sufficient to explain the behavioral data collected in our experiments.…”

“…In the following subsections, we introduce the techniques used to compute the descriptors for loudness, tempo (derived from duration), and brightness. These features are the most general and powerful descriptors for natural sounds [28]. A model capable of predicting the default foreground based on these global features would be a convenient design tool for sonification applications.…”

“…Brightness (B) is considered one of the independent perceptual dimensions for most sound categories and is particularly relevant for environmental sounds [28]. For this purpose, we use the spectral centroid [29], more specifically, the complex spectral centroid (sC) suggested by Misdariis et al [28].…”

“…For this purpose, we use the spectral centroid [29], more specifically, the complex spectral centroid (sC) suggested by Misdariis et al [28]. We used 2048 points for the FFT (FS = 44.1 kHz) and a time window of 46 ms, with a step size of 5 ms for all computations.…”