Birdvox-Full-Night: A Dataset and Benchmark for Avian Flight Call Detection

Lostanlen, Vincent; Farnsworth, Andrew; Kelling, Steve; Bello, Juan Pablo

doi:10.1109/icassp.2018.8461410

Cited by 37 publications

(106 citation statements)

References 15 publications

Supporting

Mentioning

105

Contrasting

Unclassified

Order By: Relevance

“…In this setting, the input to the system is a short audio clip, whose duration is equal to 150 ms. We represent this audio clip by a time-frequency representation E(t, f ). In the state-of-the-art model of [57], the matrix E(t, f ) contains the magnitudes in the mel-frequency spectrogram near time t and mel frequency f . The output to the system is a number y between 0 and 1, denoting the probability of presence of a sound event of interest.…”

Section: Methods Overviewmentioning

confidence: 99%

“…Yet, despite its simplicity and computational efficiency, such algorithms suffer from considerable shortcomings in detection accuracy, and may not be a reliable replacement for human inspection. In particular, a previous evaluation campaign showed that these detectors can exhibit precision and recall metrics both below 10% in a multi-sensor setting [57].…”

Section: Evidence Of Technical Bias In State-of-the-art Bioacoustic Dmentioning

confidence: 99%

“…We use the muda Python package (MUsical Data Augmentation [89]) to apply 20 randomized digital audio effects to each audio clip: four pitch shifts; four time stretchings; and four additions of background noise originating from each of all three training sensors in the cross-validation fold at hand. The choices of probability distributions and hyperparameters underlying these augmentations are identical to those of [57], and are chosen in accordance with expert knowledge about the typical vocal ranges of thrushes, warblers, and sparrows.…”

Section: May 22 2019 14/30mentioning

confidence: 99%

“…To this aim, we begin by extracting local peaks in the event detection function according to a fixed threshold τ. The baseline CNN model of [57] constrains consecutive detections to be spaced in time by a minimum lag of at least 150 ms. This constraint improved precision in the baseline CNN model without much detriment to recall, and for consistency we kept this constraint throughout our evaluation.…”

Section: May 22 2019 15/30mentioning

confidence: 99%

See 3 more Smart Citations

Robust sound event detection in bioacoustic sensor networks

et al. 2019

Self Cite

View full text Add to dashboard Cite

Bioacoustic sensors, sometimes known as autonomous recording units (ARUs), can record sounds of wildlife over long periods of time in scalable and minimally invasive ways. Deriving per-species abundance estimates from these sensors requires detection, classification, and quantification of animal vocalizations as individual acoustic events. Yet, variability in ambient noise, both over time and across sensors, hinders the reliability of current automated systems for sound event detection (SED), such as convolutional neural networks (CNN) in the time-frequency domain. In this article, we develop, benchmark, and combine several machine listening techniques to improve the generalizability of SED models across heterogeneous acoustic environments. As a case study, we consider the problem of detecting avian flight calls from a ten-hour recording of nocturnal bird migration, recorded by a network of six ARUs in the presence of heterogeneous background noise. Starting from a CNN yielding state-of-the-art accuracy on this task, we introduce two noise adaptation techniques, respectively integrating short-term (60 ms) and long-term (30 min) context. First, we apply per-channel energy normalization (PCEN) in the time-frequency domain, which applies short-term automatic gain control to every subband in the mel-frequency spectrogram. Secondly, we replace the last dense layer in the network by a context-adaptive neural network (CA-NN) layer, i.e. an affine layer whose weights are dynamically adapted at prediction time by an auxiliary network taking long-term summary statistics of spectrotemporal features as input. We show that PCEN reduces temporal overfitting across dawn vs. dusk audio clips whereas context adaptation on PCEN-based summary statistics reduces spatial overfitting across sensor locations. Moreover, combining them yields state-of-the-art results that are unmatched by artificial data augmentation alone. We release a pre-trained version of our best performing system under the name of BirdVoxDetect, a ready-to-use detector of avian flight calls in field recordings.

show abstract

Section: Methods Overviewmentioning

confidence: 99%

Section: Evidence Of Technical Bias In State-of-the-art Bioacoustic Dmentioning

confidence: 99%

Section: May 22 2019 14/30mentioning

confidence: 99%

Section: May 22 2019 15/30mentioning

confidence: 99%

See 2 more Smart Citations

Robust sound event detection in bioacoustic sensor networks

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, due to the scarcity of the labeled data, only a few studies have addressed the task of vocalization segmentation and species identification using deep learning approaches. Lostanlen et al 20 released a bird flight call detection dataset along with a CNN-based benchmark. Salamon et al 21 experimentally showed that the late fusion of scores obtained from CNN (deep learning) and a random forest classifier (shallow learning) results in better performance for the task of bird species classification from flight calls.…”

Section: Introductionmentioning

confidence: 99%

Deep metric learning for bioacoustic classification: Overcoming training data scarcity using dynamic triplet loss

Thakur

Thapar

Rajan

et al. 2019

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

This paper proposes multiscale convolutional neural network (CNN)-based deep metric learning for bioacoustic classification, under low training data conditions. The proposed CNN is characterized by the utilization of four different filter sizes at each level to analyze input feature maps. This multiscale nature helps in describing different bioacoustic events effectively: smaller filters help in learning the finer details of bioacoustic events, whereas, larger filters help in analyzing a larger context leading to global details. A dynamic triplet loss is employed in the proposed CNN architecture to learn a transformation from the input space to the embedding space, where classification is performed. The triplet loss helps in learning this transformation by analyzing three examples, referred to as triplets, at a time where intra-class distance is minimized while maximizing the inter-class separation by a dynamically increasing margin. The number of possible triplets increases cubically with the dataset size, making triplet loss more suitable than the softmax cross-entropy loss in low training data conditions. Experiments on three different publicly available datasets show that the proposed framework performs better than existing bioacoustic classification frameworks. Experimental results also confirm the superiority of the triplet loss over the cross-entropy loss in low training data conditions.

show abstract

Exploring Fine-Grained Audiovisual Categorization with the SSW60 Dataset

Horn

Qian

Wilber

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

We present a new benchmark dataset, Sapsucker Woods 60 (SSW60), for advancing research on audiovisual fine-grained categorization. While our community has made great strides in fine-grained visual categorization on images, the counterparts in audio and video finegrained categorization are relatively unexplored. To encourage advancements in this space, we have carefully constructed the SSW60 dataset to enable researchers to experiment with classifying the same set of categories in three different modalities: images, audio, and video. The dataset covers 60 species of birds and is comprised of images from existing datasets, and brand new, expert curated audio and video datasets. We thoroughly benchmark audiovisual classification performance and modality fusion experiments through the use of state-of-the-art transformer methods. Our findings show that performance of audiovisual fusion methods is better than using exclusively image or audio based methods for the task of video classification. We also present interesting modality transfer experiments, enabled by the unique construction of SSW60 to encompass three different modalities. We hope the SSW60 dataset and accompanying baselines spur research in this fascinating area.

show abstract

Birdvox-Full-Night: A Dataset and Benchmark for Avian Flight Call Detection

Cited by 37 publications

References 15 publications

Robust sound event detection in bioacoustic sensor networks

Robust sound event detection in bioacoustic sensor networks

Deep metric learning for bioacoustic classification: Overcoming training data scarcity using dynamic triplet loss

Exploring Fine-Grained Audiovisual Categorization with the SSW60 Dataset

Contact Info

Product

Resources

About