The ability of deep convolutional neural networks (CNNs) to learn discriminative spectro-temporal patterns makes them well suited to environmental sound classification. However, the relative scarcity ...of labeled data has impeded the exploitation of this family of high-capacity models. This study has two primary contributions: first, we propose a deep CNN architecture for environmental sound classification. Second, we propose the use of audio data augmentation for overcoming the problem of data scarcity and explore the influence of different augmentations on the performance of the proposed CNN architecture. Combined with data augmentation, the proposed model produces state-of-the-art results for environmental sound classification. We show that the improved performance stems from the combination of a deep, high-capacity model and an augmented training set: this combination outperforms both the proposed CNN without augmentation and a "shallow" dictionary learning model with augmentation. Finally, we examine the influence of each augmentation on the model's classification accuracy for each class, and observe that the accuracy for each class is influenced differently by each augmentation, suggesting that the performance of the model could be improved further by applying class-conditional data augmentation.
Sound event detection (SED) methods are tasked with labeling segments of audio recordings by the presence of active sound sources. SED is typically posed as a supervised machine learning problem, ...requiring strong annotations for the presence or absence of each sound source at every time instant within the recording. However, strong annotations of this type are both labor- and cost-intensive for human annotators to produce, which limits the practical scalability of SED methods. In this paper, we treat SED as a multiple instance learning (MIL) problem, where training labels are static over a short excerpt, indicating the presence or absence of sound sources but not their temporal locality. The models, however, must still produce temporally dynamic predictions, which must be aggregated (pooled) when comparing against static labels during training. To facilitate this aggregation, we develop a family of adaptive pooling operators - referred to as autopool - which smoothly interpolate between common pooling operators, such as min-, max-, or average-pooling, and automatically adapt to the characteristics of the sound sources in question. We evaluate the proposed pooling operators on three datasets, and demonstrate that in each case, the proposed methods outperform nonadaptive pooling operators for static prediction, and nearly match the performance of models trained with strong, dynamic annotations. The proposed method is evaluated in conjunction with convolutional neural networks, but can be readily applied to any differentiable model for time-series label prediction. While this paper focuses on SED applications, the proposed methods are general, and could be applied widely to MIL problems in any domain.
Per-Channel Energy Normalization: Why and How Lostanlen, Vincent; Salamon, Justin; Cartwright, Mark ...
IEEE signal processing letters,
2019-Jan., 2019-1-00, 20190101, 2019-01, Letnik:
26, Številka:
1
Journal Article
Recenzirano
Odprti dostop
In the context of automatic speech recognition and acoustic event detection, an adaptive procedure named per-channel energy normalization (PCEN) has recently shown to outperform the pointwise ...logarithm of mel-frequency spectrogram (logmelspec) as an acoustic frontend. This letter investigates the adequacy of PCEN for spectrogram-based pattern recognition in far-field noisy recordings, both from theoretical and practical standpoints. First, we apply PCEN on various datasets of natural acoustic environments and find empirically that it Gaussianizes distributions of magnitudes while decorrelating frequency bands. Second, we describe the asymptotic regimes of each component in PCEN: temporal integration, gain control, and dynamic range compression. Third, we give practical advice for adapting PCEN parameters to the temporal properties of the noise to be mitigated, the signal to be enhanced, and the choice of time-frequency representation. As it converts a large class of real-world soundscapes into additive white Gaussian noise, PCEN is a computationally efficient frontend for robust detection and classification of acoustic events in heterogeneous environments.
•Shows the suitability of a MEMS mic. for accurate acoustic sensing at type-2 level.•A consumer computing core allows for advanced DSP in-situ such as auto source ID.•Low-cost components provide an ...advanced and scalable system for acoustic sensing.
The urban sound environment of New York City (NYC) can be, amongst other things: loud, intrusive, exciting and dynamic. As indicated by the large majority of noise complaints registered with the NYC 311 information/complaints line, the urban sound environment has a profound effect on the quality of life of the city’s inhabitants. To monitor and ultimately understand these sonic environments, a process of long-term acoustic measurement and analysis is required. The traditional method of environmental acoustic monitoring utilizes short term measurement periods using expensive equipment, setup and operated by experienced and costly personnel. In this paper a different approach is proposed to this application which implements a smart, low-cost, static, acoustic sensing device based around consumer hardware. These devices can be deployed in numerous and varied urban locations for long periods of time, allowing for the collection of longitudinal urban acoustic data. The varied environmental conditions of urban settings make for a challenge in gathering calibrated sound pressure level data for prospective stakeholders. This paper details the sensors’ design, development and potential future applications, with a focus on the calibration of the devices’ Microelectromechanical systems (MEMS) microphone in order to generate reliable decibel levels at the type/class 2 level.
A considerable challenge in applying deep learning to audio classification is the scarcity of labeled data. An increasingly popular solution is to learn deep audio embeddings from large audio ...collections and use them to train shallow classifiers using small labeled datasets. Look, Listen, and Learn (L 3 -Net) is an embedding trained through self-supervised learning of audio-visual correspondence in videos as opposed to other embeddings requiring labeled data. This framework has the potential to produce powerful out-of-the-box embeddings for downstream audio classification tasks, but has a number of unexplained design choices that may impact the embeddings' behavior. In this paper we investigate how L 3 -Net design choices impact the performance of downstream audio classifiers trained with these embeddings. We show that audio-informed choices of input representation are important, and that using sufficient data for training the embedding is key. Surprisingly, we find that matching the content for training the embedding to the downstream task is not beneficial. Finally, we show that our best variant of the L 3 -Net embedding outperforms both the VGGish and SoundNet embeddings, while having fewer parameters and being trained on less data. Our implementation of the L 3 -Net embedding model as well as pre-trained models are made freely available online.
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.
The objective of this study was to evaluate consumer habits as well as the sensory perception and characteristics of farm eggs produced in Los Ríos, Chile. Data were collected from an online survey ...of 197 respondents and a sensory evaluation carried out by 30 untrained panelists of 4 types of eggs (brown-shell and blue-shell eggs acquired from family farms, free-range eggs acquired from large, industrial systems, and white-shell cage eggs from industrial, cage systems.) To evaluate differences and preferences, data were analyzed in a GLM. In addition, sensory evaluation was analyzed using principal component analysis. In accordance with the survey, 99% of the participants eat eggs (P < 0.001), 58% eat 1 to 3 eggs/wk, and 84% declared to consume eggs at home (<0.0001). Surveyed participants reported that price and size are the determining factors (31%) when purchasing eggs. Among the physical characteristic for consumers, yolk color was the most important attribute rather than white color, egg appearance, texture, flavor, or odor. In the consumer acceptability test, farm eggs (either brown or blue shell) received the most favorable sensory evaluation by the panel and were preferred to both free-range and white-shell cage eggs. Yolk color was the most influential parameter in making this difference. Brown farm eggs were predominately selected for greatest general satisfaction by participants in both the sensory evaluation (P = 0.008) and in the survey (40%; P = 0.026). There were no differences between farm eggs (brown and blue shell, P > 0.05) in the evaluated parameters. There was a consequence in the information given from surveyed consumers and the sensory panel with the yolk color.
Monitoring small, mobile organisms is crucial for science and conservation, but is technically challenging. Migratory birds are prime examples, often undertaking nocturnal movements of thousands of ...kilometres over inaccessible and inhospitable geography. Acoustic technology could facilitate widespread monitoring of nocturnal bird migration with minimal human effort. Acoustics complements existing monitoring methods by providing information about individual behaviour and species identities, something generally not possible with tools such as radar. However, the need for expert humans to review audio and identify vocalizations is a challenge to application and development of acoustic technologies.
Here, we describe an automated acoustic monitoring pipeline that combines acoustic sensors with machine listening software (BirdVoxDetect). We monitor 4 months of autumn migration in the northeastern United States with five acoustic sensors, extracting nightly estimates of nocturnal calling activity of 14 migratory species with distinctive flight calls. We examine the ability of acoustics to inform two important facets of bird migration: (1) the quantity of migrating birds aloft and (2) the migration timing of individual species. We validate these data with contemporaneous observations from Doppler radars and a large community of citizen scientists, from which we derive independent measures of migration passage and timing.
Together, acoustic and weather data produced accurate estimates of the number of actively migrating birds detected with radar. A model combining acoustic data, weather and seasonal timing explained 75% of variation in radar‐derived migration intensity. This model outperformed models that lacked acoustic data. Including acoustics in the model decreased prediction error by 33%. A model with only acoustic information outperformed a model comprising weather and date (57% vs. 48% variation explained, respectively).
Acoustics also successfully measured migration phenology: species‐specific timing estimated by acoustic sensors explained 71% of variation in timing derived from citizen science observations.
Synthesis and applications. Our results demonstrate that cost‐effective acoustic sensors can monitor bird migration at species resolution at the landscape scale and should be an integral part of management toolkits. Acoustic monitoring presents distinct advantages over radar and human observation, especially in inaccessible and inhospitable locations, and requires significantly less expense. Managers should consider using acoustic tools for monitoring avian movements and identifying and understanding dangerous situations for birds. These recommendations apply to a variety of conservation and policy applications, including mitigating the impacts of light pollution, siting energy infrastructure (e.g. wind turbines) and reducing collisions with structures and aircraft.
Our results demonstrate that cost‐effective acoustic sensors can monitor bird migration at species resolution at the landscape scale and should be an integral part of management toolkits. Acoustic monitoring presents distinct advantages over radar and human observation, especially in inaccessible and inhospitable locations, and requires significantly less expense. Managers should consider using acoustic tools for monitoring avian movements and identifying and understanding dangerous situations for birds. These recommendations apply to a variety of conservation and policy applications, including mitigating the impacts of light pollution, siting energy infrastructure (e.g. wind turbines) and reducing collisions with structures and aircraft.
Automatic classification of animal vocalizations has great potential to enhance the monitoring of species movements and behaviors. This is particularly true for monitoring nocturnal bird migration, ...where automated classification of migrants' flight calls could yield new biological insights and conservation applications for birds that vocalize during migration. In this paper we investigate the automatic classification of bird species from flight calls, and in particular the relationship between two different problem formulations commonly found in the literature: classifying a short clip containing one of a fixed set of known species (N-class problem) and the continuous monitoring problem, the latter of which is relevant to migration monitoring. We implemented a state-of-the-art audio classification model based on unsupervised feature learning and evaluated it on three novel datasets, one for studying the N-class problem including over 5000 flight calls from 43 different species, and two realistic datasets for studying the monitoring scenario comprising hundreds of thousands of audio clips that were compiled by means of remote acoustic sensors deployed in the field during two migration seasons. We show that the model achieves high accuracy when classifying a clip to one of N known species, even for a large number of species. In contrast, the model does not perform as well in the continuous monitoring case. Through a detailed error analysis (that included full expert review of false positives and negatives) we show the model is confounded by varying background noise conditions and previously unseen vocalizations. We also show that the model needs to be parameterized and benchmarked differently for the continuous monitoring scenario. Finally, we show that despite the reduced performance, given the right conditions the model can still characterize the migration pattern of a specific species. The paper concludes with directions for future research.
Noise pollution is one of the topmost quality of life issues for urban residents in the United States. Continued exposure to high levels of noise has proven effects on health, including acute effects ...such as sleep disruption, and long-term effects such as hypertension, heart disease, and hearing loss. To investigate and ultimately aid in the mitigation of urban noise, a network of 55 sensor nodes has been deployed across New York City for over two years, collecting sound pressure level (SPL) and audio data. This network has cumulatively amassed over 75 years of calibrated, high-resolution SPL measurements and 35 years of audio data. In addition, high frequency telemetry data have been collected that provides an indication of a sensors' health. These telemetry data were analyzed over an 18-month period across 31 of the sensors. It has been used to develop a prototype model for pre-failure detection which has the ability to identify sensors in a prefail state 69.1% of the time. The entire network infrastructure is outlined, including the operation of the sensors, followed by an analysis of its data yield and the development of the fault detection approach and the future system integration plans for this.