Automated detection of Hainan gibbon calls for passive acoustic monitoring

Abstract

Extracting species calls from passive acoustic recordings is a common preliminary step to ecological analysis. For many species, particularly those occupying noisy, acoustically variable habitats, the call extraction process continues to be largely manual, a time-consuming and increasingly unsustainable process. Deep neural networks have been shown to offer excellent performance across a range of acoustic classification applications, but are relatively underused in ecology. We describe the steps involved in developing an automated classifier for a passive acoustic monitoring project, using the identification of calls of the Hainan gibbon Nomascus hainanus, one of the world's rarest mammal species, as a case study. This includes preprocessing-selecting a temporal resolution, windowing and annotation; data augmentation; processing-choosing and fitting appropriate neural network models; and post-processing-linking model predictions to replace, or more likely facilitate, manual labelling. Our best model converted acoustic recordings into spectrogram images on the mel frequency scale, using these to train a convolutional neural network. Model predictions were highly accurate, with per-second false positive and false negative rates of 1.5% and 22.3%. Nearly all false negatives were at the fringes of calls, adjacent to segments where the call was correctly identified, so that very few calls were missed altogether. A post-processing step identifying intervals of repeated calling reduced an 8-h recording to, on average, 22 min for manual processing, and did not miss any calling bouts over 72 h of test recordings. Gibbon calling bouts were detected regularly in multi-month recordings from all selected survey points within Bawangling National Nature Reserve, Hainan. We demonstrate that passive acoustic monitoring incorporating an automated classifier represents an effective tool for remote detection of one of the world's rarest and most threatened species. Our study highlights the viability of using neural networks to automate or greatly assist the manual labelling of data collected by passive acoustic monitoring projects. We emphasize that model development and implementation be informed and guided by ecological objectives, and increase accessibility of these tools with a series of notebooks that allow users to build and deploy their own acoustic classifiers.

Novel Technologies and Their Application for Protected Area Management: A Supporting Approach in Biodiversity Monitoring

“…Sound is recognized as a common means of communication in insects, fish, birds, squamates, and mammals [98,100]. Call count censusing has long been a standard practice to identify community assemblages [101,102]. Initially conducted with expensive, cumbersome equipment, census techniques using recorders now allow ecologists to document a wide diversity of species at a far lower cost than continual deployment of field crews [98].…”

“…Expert-based field identification may compare favorably to findings generated from remote microphone arrays linked to species recognition algorithms [108]. Yet, surveys relying on human skill for identification of species are prone to error due to imperfect species detection, confirmation bias, or listener fatigue [102,103,119,122]. Lack of objective classification is especially challenging when a reviewer is charged with identifying rare or unknown species, with animals that are known to employ mimicry, or in complex soundscapes [104,111].…”

“…Analysis of soundscape profiles can be facilitated through reduction of background noise [104,109]. Incorporating species time of arrival or activity into a survey using fixed-point microphone arrays can be an approach to reduce bias [102,114]. Through application of sound filters, automated programs can eliminate sections of uninformative data, facilitating verification of acoustic signals by a reviewer [117].…”

“…Important limitations persist for auditory species identification. Use of automated computer recognition of animal calls is currently underutilized [102]. For effective ML, hundreds of labeled sound records are required [115,120].…”

BioCPPNet: Automatic Bioacoustic Source Separation with Deep Neural Networks

“…While the human speech separation problem is a competitive area of work, the bioacoustic CPP has received comparatively less attention, as current bioacoustic research often emphasizes other ML-based tasks such as automated detection and classification of bioacoustic sounds [30][31][32] . However, recent work has implemented both semi-classical and deep ML-based approaches to address bioacoustic source separation, employing time domain and TFR-based algorithms.…”