Giovanni Brambilla

Urban environments are characterized by a complex interplay of various sound sources, which significantly influence the overall soundscape quality. This study presents a methodology that combines the intermittency ratio (IR) metric for acoustic event detection with deep learning (DL) techniques for the classification of sound sources associated with these events. The aim is to provide an automated tool for detecting and categorizing polyphonic acoustic events, thereby enhancing our ability to assess and manage environmental noise. Using a dataset collected in the city center of Barcelona, our results demonstrate the effectiveness of the IR metric in successfully detecting events from diverse categories. Specifically, the IR captures the temporal variations of sound pressure levels due to significant noise events, enabling their detection but not providing information on the associated sound sources. To fill this weakness, the DL-based classification system, which uses a MobileNet convolutional neural network, shows promise in identifying foreground sound sources. Our findings highlight the potential of DL techniques to automate the classification of sound sources, providing valuable insights into the acoustic environment. The proposed methodology of combining the two above techniques represents a step forward in automating acoustic event detection and classification in urban soundscapes and providing important information to manage noise mitigation actions.

Noise annoyance depends not only on sound energy, but also on other features, such as those in its spectrum (e.g., low frequency and/or tonal components), and, over time, amplitude fluctuations, such as those observed in road, rail, or aircraft noise passages. The larger these fluctuations, the more annoying a sound is generally perceived. Many algorithms have been implemented to quantify these fluctuations and identify noise events, either by looking at transients in the sound level time history, such as exceedances above a fixed or time adaptive threshold, or focusing on the hearing perception process of such events. In this paper, four criteria to detect sound were applied to the acoustic monitoring data collected in two urban areas, namely Andorra la Vella, Principality of Andorra, and Milan, Italy. At each site, the 1 s A-weighted short L_Aeq,1s time history, 10 min long, was available for each hour from 8:00 a.m. to 7:00 p.m. The resulting 92-time histories cover a reasonable range of urban environmental noise time patterns. The considered criteria to detect noise events are based on: (i) noise levels exceeding by +3 dB the continuous equivalent level L_AeqT referred to the measurement time (T), criteria used in the definition of the Intermittency Ratio (IR) to detect noise events; (ii) noise levels exceeding by +3 dB the running continuous equivalent noise level; (iii) noise levels exceeding by +10 dB the 50th noise level percentile; (iv) progressive positive increments of noise levels greater than 10 dB from the event start time. Algorithms (iii) and (iv) appear suitable for notice-event detection; that is, those that (for their features) are clearly perceived and potentially annoy exposed people. The noise events detected by the above four algorithms were also evaluated by the available anomalous noise event detection (ANED) procedure to classify them as produced by road traffic noise or something else. Moreover, the assessment of the sonic environment by the Harmonica index was correlated with the single event level (SEL) of each event detected by the four algorithms. The threshold value of 8 for the Harmonica index, separating the “noisy” from the “very noisy” environments, corresponds to lower SEL levels for notice-events as identified by (iii) and (iv) algorithms (about 88–89 dB(A)) against those identified by (i) and (ii) criteria (92 dB(A)).

We investigated the statistical properties of the sounds recorded at a site located in the Parco Nord of Milan, Italy, characterized by wooded land rich in biodiversity and exposed to different sources and types of anthropogenic disturbances with the aim of deriving information on its environmental quality in terms of biophonic presence and athropic disturbance. A time series of eco-acosutic indices were determined for 616 audio files recorded from 30 April 2019 (5:00 p.m.) to 3 May 2019 (5:00 a.m.) with a 1-min duration followed by a 5-min pause (10 recordings per hour). In the present study, the following indices were computed: the Acoustic Complexity Index (ACI), Acoustic Diversity Index (ADI), Acoustic Evenness Index (AEI), Bio-acoustic Index (BI), Acoustic Entropy Index (H), Normalized Difference Soundscape Index (NSDI) and Dynamic Spectral Centroid (DSC). Cluster analysis performed on the corresponding time series yielded a dimensional reduction from seven down to three. The results show a clear separation of the eco-acoustic indices into two clusters, reflecting the different dynamics and diversity behaviour throughout the recordings. A post-processing aural survey was also performed, aiming at determining biophonic activities (mainly avian vocalization and other animals), the characteristics of technophonies sources (mainly road traffic noise and airplane fly-overs), human presence (voices and steps) and geophonies (rain and wind). The statistical analysis proved to be a robust tool due to the good matching obtained with the aural survey outcomes. The overall quality of the Parco Nord phonic activity was found to be low. Notwithstanding the presence of avian species, highlighted by the characteristic dawn chorus, both clusters revealed low “scores” of NDSI and DSC indices heavily influenced by road traffic sources. This study represents the first step toward the realization of maps of eco-acoustic indices for the long-term monitoring of fragile habitats.

In the last decades, the soundscape approach has attracted the attention of architects and urban planners, leading them to incorporate the acoustic features into the enjoyment of their creations. One of the key aspects for an appreciated urban environment is to match the expectations of the users. In this study, the matching of the waterfront soundscape with the users’ expectations is evaluated by laboratory tests using semantic differential scales applied to reproduced virtual scenarios obtained adding different water sound pressure levels (SPLs) to the original in-situ setting. The tests were carried out by an immersive virtual reality (IVR) device, using 360° videos and spatial audio recorded in two sites of the waterfront in Naples, Italy. The scenarios were presented to the participants according to three experimental protocols, namely audio-only (A), video-only (V), and simultaneous audio-video (AV) reproduction. The examined different acoustic scenarios were the original one recorded in situ and others obtained adding seawater sounds at SPL increments of 5 dB. The results show that all the scenarios with water sounds added are rated more pleasant than the original one for the audio-only scenario. When video and audio are displayed simultaneously, two scenarios are more pleasant than the original one, likely because there is a need for coherence between the water sound SPL heard and the visible noise sources. Sounds coherent with the type of shore show a higher matching with expectations and pleasantness appraisals, rather than those that are uncoherent with the layout scenario.

The characterization of environmental quality and the detection of the first sign of environmental stress, with reference to human intrusion, is currently a very important goal to prevent further environmental degradation, and consequently habitat destruction, in order to take appropriate preservation measures. Besides the traditional field observation and satellite remote sensing, geophonic and/or biophonic sounds have been proposed as potential indicators of terrestrial and aquatic settings’ status. In this work, we analyze a series of short audio-recordings taken in urban parks and bushes characterized by the presence of different human-generated-noise and species abundance. This study aims to propose a tool devoted to the investigation of urban and natural environments in a context with different soundscape qualities, such as, for example, those that can be found in urban parks. The analysis shows the ways in which it is possible to distinguish among different habitats by the use of a combination of different acoustic and sound ecology indices.

The urban environmental planning, a fundamental dynamic process for cities’ sustainability, could benefit from the soundscape approach, dealing with the perception of the acoustic environment in which sound is considered as a resource rather than a waste (noise). Noise and soundscape maps are useful tools for planning mitigation actions and for communication with citizens. Both mappings can benefit from crowdsourcing and participatory sound monitoring that has been made possible due to the large use of internet connections and mobile devices with dedicated apps. This paper is a “scoping review” to provide an overview of the potential, benefits, and drawbacks of participatory noise monitoring in noise and soundscape mapping applications, while also referring to metrological aspects. Gathering perceptual data on soundscapes by using digital questionnaires will likely be more commonly used than printed questionnaires; thus, the main differences between the experimental protocols concern the measurement of acoustic data. The authors propose to classify experimental protocols for in-field soundscape surveys into three types (GUIDE, MONITOR, and SMART) to be selected according to the survey’s objectives and the territorial extension. The main future developments are expected to be related to progress in smartphone hardware and software, to the growth of social networks data analysis, as well as to the implementation of machine learning techniques.

Noise energetic indicators, like L_den, show good correlations with long term annoyance, but should be supplemented by other parameters describing the sound fluctuations, which are very common in urban areas and negatively impact noise annoyance. Thus, in this paper, the hourly values of continuous equivalent level L_Aeqh and the intermittency ratio (IR) were both considered to describe the urban road traffic noise, monitored in 90 sites in the city of Milan and covering different types of road, from motorways to local roads. The noise data have been processed by clustering methods to detect similarities and to figure out a criterion to classify the urban sites taking into account both equivalent noise levels and road traffic noise events. Two clusters were obtained and, considering the cluster membership of each site, the decimal logarithm of the day-time (06:00–22:00) traffic flow was used to associate each new road with the clusters. In particular, roads with average day-time hourly traffic flow ≥1900 vehicles/hour were associated with the cluster with high traffic flow. The described methodology could be fruitfully applied on road traffic noise data in other cities.

Human hearing adapts to steady signals, but remains very sensitive to fluctuations as well as to prominent, salient noise events. The higher these fluctuations are, the more annoying a sound is possibly perceived. To quantify these fluctuations, descriptors have been proposed in the literature and, among these, the intermittency ratio (IR) has been formulated to quantify the eventfulness of an exposure from transportation noise. This paper deals with the application of IR to urban road traffic noise data, collected in terms of 1 s A-weighted sound pressure level (SPL), without being attended, monitored continuously for 24 h in 90 sites in the city of Milan. IR was computed on each hourly data of the 251 time series available (lasting 24 h each), including different types of roads, from motorways to local roads with low traffic flow. The obtained hourly IR values have been processed by clustering methods to extract the most significant temporal pattern features of IR in order to figure out a criterion to classify the urban sites taking into account road traffic noise events, which potentially increase annoyance. Two clusters have been obtained and a “non-acoustic” parameter x, determined by combination of the traffic flow rate in three hourly intervals, has allowed to associate each site with the cluster membership. The described methodology could be fruitfully applied on road traffic noise data in other cities. Moreover, to have a more detailed characterization of noise exposure, IR, describing SPL short-term temporal variations, has proved to be a useful supplementary metric accompanying L_Aeq, which is limited to measure the energy content of the noise exposure.