No abstract

A total of 147 days spread over 4 years were recorded by a stereophonic sonobuoy set up in the Mediterranean sea, near the coast of Toulon, south of France. These recordings were analyzed in the scope of studying sperm whales (Physeter macrocephalus) and the impact anthropic noises may have on this species. With the use of a novel approach, which combines the use of a stereophonic antenna with a neural network, 226 sperm whales’ passages have been automatically detected in an effective range of 32 km. This dataset was then used to analyze the sperm whales’ abundance, the background noise, the influence of the background noise on the acoustic presence, and the animals’ size. The results show that sperm whales are present all year round in groups of 1–9 individuals, especially during the daytime. The estimated density is 1.69 whales/1000 km$$^2$$ 2 . Animals were also less frequent during periods with an increased background noise due to ferries. The animal size distribution revealed the recorded sperm whales were distributed in length from about 7 to 15.5 m, and lonely whales are larger, while groups of two are composed of juvenile and mid-sized animals.

In this paper, we study the difficulties of domain transfer when training deep learning models, on a specific task that is orca vocalization detection. Deep learning appears to be an answer to many sound recognition tasks in human speech analysis as well as in bioacoustics. This method allows to learn from large amounts of data, and find the best scoring way to discriminate between classes (e.g. orca vocalization and other sounds). However, to learn the perfect data representation and discrimination boundaries, all possible data configurations need to be processed. This causes problems when those configurations are ever changing (e.g. in our experiment, a change in the recording system happened to considerably disturb our previously well performing model). We thus explore approaches to compensate on the difficulties faced with domain transfer, with two convolutionnal neural networks (CNN) architectures, one that works in the time-frequency domain, and one that works directly on the time domain.

We present an analysis of fin whale (Balaenoptera physalus) songs on passive acoustic recordings from the Pelagos Sanctuary (Western Mediterranean Basin). The recordings were gathered between 2008 and 2018 using 2 different hydrophone stations. We show how 20 Hz fin whale pulses can be automatically detected using a low complexity convolutional neural network (CNN) despite data variability (different recording devices exposed to diverse noises). The pulses were further classified into the two categories described in past studies and inter pulse intervals (IPI) were measured. The results confirm previous observations on the local relationship between pulse type and IPI with substantially more data. Furthermore we show inter-annual shifts in IPI and an intra-annual trend in pulse center frequency. This study provides new elements of comparison for the understanding of long term fin whale song trends worldwide.

Killer whales (Orcinus orca) can produce 3 types of signals: clicks, whistles and vocalizations. This study focuses on Orca vocalizations from northern Vancouver Island (Hanson Island) where the NGO Orcalab developed a multi-hydrophone recording station to study Orcas. The acoustic station is composed of 5 hydrophones and extends over 50 km 2 of ocean. Since 2015 we are continuously streaming the hydrophone signals to our laboratory in Toulon, France, yielding nearly 50 TB of synchronous multichannel recordings. In previous work, we trained a Convolutional Neural Network (CNN) to detect Orca vocalizations, using transfer learning from a bird activity dataset. Here, for each detected vocalization, we estimate the pitch contour (fundamental frequency). Finally, we cluster vocalizations by features describing the pitch contour. While preliminary, our results demonstrate a possible route towards automatic Orca call type classification. Furthermore, they can be linked to the presence of particular Orca pods in the area according to the classification of their call types. A large-scale call type classification would allow new insights on phonotactics and ethoacoustics of endangered Orca populations in the face of increasing anthropic pressure.

The study of non-human animals' communication systems generally relies on the transcription of vocal sequences using a finite set of discrete units. This set is referred to as a vocal repertoire, which is specific to a species or a sub-group of a species. When conducted by human experts, the formal description of vocal repertoires can be laborious and / or biased. This motivates a computerised assistance for this procedure, for which machine learning algorithms represent a good opportunity. Unsupervised clustering algorithms are suited for grouping close points together, provided a relevant representation. This paper therefore studies a new method for encoding vocalisations, allowing for automatic clustering to alleviate vocal repertoire characterisation. Borrowing from deep representation learning, we use a convolutional auto-encoder network to learn an abstract representation of vocalisations. We report on the quality of the learnt representation, as well as of state of the art methods, by quantifying their agreement with expert labelled vocalisation types from 7 datasets of other studies across 6 species (birds and marine mammals). With this benchmark, we demonstrate that using auto-encoders improves the relevance of vocalisation representation which serves repertoire characterisation, and so using a very limited amount of settings. We also publish a Python package for the bioacoustic community to train their own vocalisation auto-encoders or use a pretrained encoder to browse vocal repertoires and ease unit wise annotation.

No abstract

Killer whales (Orcinus orca) can produce 3 types of signals: clicks, whistles and vocalizations. This study focuses on Orca vocalizations from northern Vancouver Island (Hanson Island) where the NGO Orcalab developed a multi-hydrophone recording station to study Orcas. The acoustic station is composed of 5 hydrophones and extends over 50 km 2 of ocean. Since 2015 we are continuously streaming the hydrophone signals to our laboratory in Toulon, France, yielding nearly 50 TB of synchronous multichannel recordings. In previous work, we trained a Convolutional Neural Network (CNN) to detect Orca vocalizations, using transfer learning from a bird activity dataset. Here, for each detected vocalization, we estimate the pitch contour (fundamental frequency). Finally, we cluster vocalizations by features describing the pitch contour. While preliminary, our results demonstrate a possible route towards automatic Orca call type classification. Furthermore, they can be linked to the presence of particular Orca pods in the area according to the classification of their call types. A large-scale call type classification would allow new insights on phonotactics and ethoacoustics of endangered Orca populations in the face of increasing anthropic pressure.