Methods - Soundscape Ecology: Principles, Patterns, Methods and Applications - page 256

The spectrogram is the visual representation of the sound in a space where the
horizontal dimension corresponds to time and the vertical dimension represents the
frequency (or pitch), with higher sounds shown higher on the display. The color of
the spectrogram represents the magnitude of the signal.
A field record offers a great variety of recording devices that can sample at
different rates and on different types of digital memories. Single microphones, sets
of adjacent microphones, and arrays of microphones are some of the strategies
adopted to collect sound from the environment.
The main goal of soundscape ecology is to process data to extract the emerging
patterns in terms of complexity/information. A sound pressure measurement and a
spectral-frequency analysis are the two distinct approaches to day adopted. The
sound pressure measurement, adopted by engineering and acousticians, is obtained
with at least three different types of weighting to correct the different sensitivity of
human hear to frequencies. Equivalent continuous noise level and sound exposure
level are two examples of this approach applied in the study of sound/noise
transmission across a medium.
The spectral analysis adopted by bioacousticians offers at the moment very few
metrics to evaluate the complexity of the sonic environment. The acoustic entropy
index (H), median of amplitude envelope (M), acoustic richness (AR), acoustic
dissimilarity index (D), the acoustic complexity index (ACI), and ACI evenness are
some of the available metrics.
The automated recognition of environmental sounds has been pursued by
bioacousticians. Most animal vocalizations represent acoustic signatures that can
be recognized and classified. Today, computing facilities and new algorithms allow
a satisfactory processing of sound data, although several problems still remain,
especially when long-time recordings introduce unwanted sounds that mask the
biophonies.
Acoustic diversity as results from a spectral analysis represents a significant
component of biodiversity because this diversity represent some functional traits.
Several softwares are today offered for processing sounds. Some software is
dedicated to sound editing, such as CoolEdit, Gold Wave, Spectrafoo, Amadeus
Pro, Free Audio Editor, WavePad Sound editor, Sound Studio 2; other software is
dedicated to bioacoustic analysis such as Raven, Song Scope, Avisoft, Wildspectra,
Canary, Spectra Pro, Syrinx, Sonobat, Pamguard, Soundruler, Wavesurfer,
SoundscapeMeter, Sound Analysis Pro, SeaWave, and Ishmael, and finally some
software is dedicated to the physical analysis of the spectral representation of
sounds such as Spectrogram, Syrinx pc, Osprey, Signalize, Sigview, SpectraPlus.
Because of the great amount of data that automated recording stations can
collect, it is urgent to create data banks easily browsed. A new software, Pumilio,
is proposed characterized by quick browsing and by flexible parallel computing.
The soundwalking method is suggested as an efficient approach to soundscape
description from a psychological point of view. The procedure to describe the sonic
environment also represents an important educational tool.