Scaper: A library for soundscape synthesis and augmentation

Salamon, Justin; MacConnell, Duncan; Cartwright, Mark; Li, Peter; Bello, Juan Pablo

doi:10.1109/waspaa.2017.8170052

Cited by 147 publications

(128 citation statements)

References 15 publications

Supporting

Mentioning

126

Contrasting

Unclassified

Order By: Relevance

“…Finally, each file in these datasets has a sound present, whereas urban soundscapes contain many periods of silence or geophonic sounds, two important states which are not present in UrbanSound and UrbanSound8k. Due to these factors, these datasets are unsuitable for the purpose of this research project, although recent work has overcome a few of these shortcoming through the annual Detection and Classification of Acoustic Scenes and Events challenges, and using synthesised soundscape data (Mesaros et al., ; Salamon, MacConnell, Cartwright, Li, & Bello, ). This highlights the need for an internationally coordinated effort to create a consistently labelled audio dataset from cities to support the development of automated urban environmental assessment systems with international application.…”

Section: Discussionmentioning

confidence: 99%

“…Acoustic methods are commonly used to monitor bat populations using passive ultrasonic recorders meaning bat researchers and conservationists are faced with the challenge of extracting meaningful information from large volumes of audio data. The development of automated methods for measuring bat calls in ultrasonic data has focused to date on the identification of bat species calls and many algorithms are proprietary (e.g., Szewczak, 2010;Wildlife Acoustics, 2017 (Mesaros et al, 2017;Salamon, MacConnell, Cartwright, Li, & Bello, 2017). This highlights the need for an internationally coordinated effort to create a consistently labelled audio dataset from cities to support the development of automated urban environmental assessment systems with international application.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

CityNet—Deep learning tools for urban ecoacoustic assessment

et al. 2018

View full text Add to dashboard Cite

Cities support unique and valuable ecological communities, but understanding urban wildlife is limited due to the difficulties of assessing biodiversity. Ecoacoustic surveying is a useful way of assessing habitats, where biotic sound measured from audio recordings is used as a proxy for population abundance and/or activity. However, existing algorithms systematically over and underestimate measures of biotic activity in the presence of typical urban non‐biotic sounds in recordings. We develop CityNet, a deep learning system using convolutional neural networks (CNNs), to measure audible biotic (CityBioNet) and anthropogenic (CityAnthroNet) acoustic activity in cities. The CNNs were trained on a large dataset of annotated audio recordings collected across Greater London, UK. Using a held‐out test dataset, we compare the precision and recall of CityBioNet and CityAnthroNet separately to the best available alternative algorithms: four Acoustic Indices: Acoustic Complexity Index, Acoustic Diversity Index, Bioacoustic Index, and Normalised Difference Soundscape Index, and a state‐of‐the‐art bird call detection CNN (bulbul). We also compare the effect of non‐biotic sounds on the predictions of CityBioNet and bulbul. Finally we apply CityNet to describe acoustic patterns of the urban soundscape in two sites along an urbanisation gradient. CityBioNet was the best performing algorithm for measuring biotic activity in terms of precision and recall, followed by bulbul, whereas the Acoustic Indices performed worst. CityAnthroNet outperformed the Normalised Difference Soundscape Index, but by a smaller margin than CityBioNet achieved against the competing algorithms. The CityBioNet predictions were impacted by mechanical sounds, whereas air traffic and wind sounds influenced the bulbul predictions. Across an urbanisation gradient, we show that CityNet produced realistic daily patterns of biotic and anthropogenic acoustic activity from real‐world urban audio data. Using CityNet, it is possible to automatically measure biotic and anthropogenic acoustic activity in cities from audio recordings. If embedded within an autonomous sensing system, CityNet could produce environmental data for cites at large‐scales and facilitate investigation of the impacts of anthropogenic activities on wildlife. The algorithms, code and pretrained models are made freely available in combination with two expert‐annotated urban audio datasets to facilitate automated environmental surveillance in cities.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

CityNet—Deep learning tools for urban ecoacoustic assessment

et al. 2018

View full text Add to dashboard Cite

show abstract

“…The URBAN-SED dataset is generated by the Scaper soundscape synthesis library [17]. The dataset containing 10 sound event classes is divided into 3 parts: a strongly-labeled training set (6000 clips), a strongly-labeled validation set (2000 clips) and a strongly-labeled test set (2000 clips).…”

Section: Urban-sedmentioning

confidence: 99%

Multi-Branch Learning for Weakly-Labeled Sound Event Detection

Huang

Wang

Lin

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

There are two sub-tasks implied in the weakly-supervised SED: audio tagging and event boundary detection. Current methods which combine multi-task learning with SED requires annotations both for these two sub-tasks. Since there are only annotations for audio tagging available in weaklysupervised SED, we design multiple branches with different learning purposes instead of pursuing multiple tasks. Similar to multiple tasks, multiple different learning purposes can also prevent the common feature which the multiple branches share from overfitting to any one of the learning purposes. We design these multiple different learning purposes based on combinations of different MIL strategies and different pooling methods. Experiments on the DCASE 2018 Task 4 dataset and the URBAN-SED dataset both show that our method achieves competitive performance.

show abstract

“…Second, to evaluate the influence of a soundscape's acoustic characteristics on its annotation quality, we need to control the types of sound events that occur in a soundscape and the degree to which they overlap. To achieve this, we generated the soundscapes using Scaper 2 [38], an open-source library we developed to probabilistically synthesize soundscapes from a collection of isolated sound-event recordings. Scaper provides high-level, probabilistic control over the number and types of sound events, their start times, durations, and loudness with respect to a "background" track.…”

Section: Scaper: a Soundscape Synthesizermentioning

confidence: 99%

“…In the process of answering these questions, we also raise new questions regarding the definition of ground-truth annotations and the sufficiency of a CSCW-based approach which aggregates audio annotations from a population of users. Finally, our contribution also include Scaper [38], an open-source tool for synthesizing soundscapes and CrowdCurio Audio-Annotator, a open-source, web-based tool for crowdsourced audio annotation.…”

Section: Introductionmentioning

confidence: 99%

Seeing Sound

Cartwright

Seals

Salamon

et al. 2017

Proc. ACM Hum.-Comput. Interact.

Self Cite

View full text Add to dashboard Cite

Audio annotation is key to developing machine-listening systems; yet, effective ways to accurately and rapidly obtain crowdsourced audio annotations is understudied. In this work, we seek to quantify the reliability/redundancy trade-off in crowdsourced soundscape annotation, investigate how visualizations affect accuracy and efficiency, and characterize how performance varies as a function of audio characteristics. Using a controlled experiment, we varied sound visualizations and the complexity of soundscapes presented to human annotators. Results show that more complex audio scenes result in lower annotator agreement, and spectrogram visualizations are superior in producing higher quality annotations at lower cost of time and human labor. We also found recall is more affected than precision by soundscape complexity, and mistakes can be often attributed to certain sound event characteristics. These findings have implications not only for how we should design annotation tasks and interfaces for audio data, but also how we train and evaluate machine-listening systems.

show abstract

Scaper: A library for soundscape synthesis and augmentation

Cited by 147 publications

References 15 publications

CityNet—Deep learning tools for urban ecoacoustic assessment

CityNet—Deep learning tools for urban ecoacoustic assessment

Multi-Branch Learning for Weakly-Labeled Sound Event Detection

Seeing Sound

Contact Info

Product

Resources

About