Learning With Out-of-Distribution Data for Audio Classification

Iqbal, Turab; Cao, Yin; Kong, Qiuqiang; Plumbley, Mark D.; Wang, Wenwu

doi:10.1109/icassp40776.2020.9054444

Cited by 19 publications

(8 citation statements)

References 11 publications

(26 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We train a VGG classifier on the proper training data of the iD dataset to classify the classes in the set, using data augmentation to learn iD equivariance. Fifteen datapoints from each class in the iD training set are held out as calibration data for the set (this is the same setting as Iqbal et al (2020)'s validation set in FSD).…”

Section: Results On Audio Datamentioning

confidence: 99%

iDECODe: In-Distribution Equivariance for Conformal Out-of-Distribution Detection

Kaur

Jha

Roy

et al. 2022

AAAI

View full text Add to dashboard Cite

Machine learning methods such as deep neural networks (DNNs), despite their success across different domains, are known to often generate incorrect predictions with high confidence on inputs outside their training distribution. The deployment of DNNs in safety-critical domains requires detection of out-of-distribution (OOD) data so that DNNs can abstain from making predictions on those. A number of methods have been recently developed for OOD detection, but there is still room for improvement. We propose the new method iDECODe, leveraging in-distribution equivariance for conformal OOD detection. It relies on a novel base non-conformity measure and a new aggregation method, used in the inductive conformal anomaly detection framework, thereby guaranteeing a bounded false detection rate. We demonstrate the efficacy of iDECODe by experiments on image and audio datasets, obtaining state-of-the-art results. We also show that iDECODe can detect adversarial examples. Code, pre-trained models, and data are available at https://github.com/ramneetk/iDECODe.

show abstract

Section: Results On Audio Datamentioning

confidence: 99%

iDECODe: In-Distribution Equivariance for Conformal Out-of-Distribution Detection

Kaur

Jha

Roy

et al. 2022

AAAI

View full text Add to dashboard Cite

show abstract

“…In [10], the authors worked on a dataset that was manipulated with some samples that did not belong to any class because of a labeling error. They proved that using those OOD (out of distribution) samples by separating them from the dataset can affect the training of the model in an effective way.…”

Section: Criteria and Alternatives Selectionmentioning

confidence: 99%

Intelligent Audio Signal Processing for Detecting Rainforest Species Using Deep Learning

Kumar¹,

Gupta²,

Ahmed³

et al. 2022

Intelligent Automation &Amp; Soft Computing

View full text Add to dashboard Cite

Hearing a species in a tropical rainforest is much easier than seeing them. If someone is in the forest, he might not be able to look around and see every type of bird and frog that are there but they can be heard. A forest ranger might know what to do in these situations and he/she might be an expert in recognizing the different type of insects and dangerous species that are out there in the forest but if a common person travels to a rain forest for an adventure, he might not even know how to recognize these species, let alone taking suitable action against them. In this work, a model is built that can take audio signal as input, perform intelligent signal processing for extracting features and patterns, and output which type of species is present in the audio signal. The model works end to end and can work on raw input and a pipeline is also created to perform all the preprocessing steps on the raw input. In this work, different types of neural network architectures based on Long Short Term Memory (LSTM) and Convolution Neural Network (CNN) are tested. Both are showing reliable performance, CNN shows an accuracy of 95.62% and Log Loss of 0.21 while LSTM shows an accuracy of 93.12% and Log Loss of 0.17. Based on these results, it is shown that CNN performs better than LSTM in terms of accuracy while LSTM performs better than CNN in terms of Log Loss. Further, both of these models are combined to achieve high accuracy and low Log Loss. A combination of both LSTM and CNN shows an accuracy of 97.12% and a Log Loss of 0.16.

show abstract

“…In SER, label sets in not-small datasets are inherently noisy due to reasons like sub-optimality of automatic methods used in the creation, or the difficulty of annotating audio-especially without visual cues, with large vocabularies, and because the annotation process is, sometimes, inherently subjective and ambiguous. Consequently, recent works have shown the efficacy of label noise treatment in large datasets such as AudioSet [41,85] and mid-size datasets [32,86,87].…”

Section: A Characteristicsmentioning

confidence: 99%

“…DenseNet-121. DenseNets are reported to outperform ResNets for image recognition [121], and have been recently used for SET [86,87].…”

Section: A Evaluationmentioning

confidence: 99%

FSD50K: An Open Dataset of Human-Labeled Sound Events

Fonseca¹,

Favory²,

Pons³

et al. 2020

Preprint

View full text Add to dashboard Cite

Most existing datasets for sound event recognition (SER) are relatively small and/or domain-specific, with the exception of AudioSet, based on a massive amount of audio tracks from YouTube videos and encompassing over 500 classes of everyday sounds. However, AudioSet is not an open datasetits release consists of pre-computed audio features (instead of waveforms), which limits the adoption of some SER methods. Downloading the original audio tracks is also problematic due to constituent YouTube videos gradually disappearing and usage rights issues, which casts doubts over the suitability of this resource for systems' benchmarking. To provide an alternative benchmark dataset and thus foster SER research, we introduce FSD50K, an open dataset containing over 51k audio clips totalling over 100h of audio manually labeled using 200 classes drawn from the AudioSet Ontology. The audio clips are licensed under Creative Commons licenses, making the dataset freely distributable (including waveforms). We provide a detailed description of the FSD50K creation process, tailored to the particularities of Freesound data, including challenges encountered and solutions adopted. We include a comprehensive dataset characterization along with discussion of limitations and key factors to allow its audio-informed usage. Finally, we conduct sound event classification experiments to provide baseline systems as well as insight on the main factors to consider when splitting Freesound audio data for SER. Our goal is to develop a dataset to be widely adopted by the community as a new open benchmark for SER research.

show abstract

Learning With Out-of-Distribution Data for Audio Classification

Cited by 19 publications

References 11 publications

iDECODe: In-Distribution Equivariance for Conformal Out-of-Distribution Detection

iDECODe: In-Distribution Equivariance for Conformal Out-of-Distribution Detection

Intelligent Audio Signal Processing for Detecting Rainforest Species Using Deep Learning

FSD50K: An Open Dataset of Human-Labeled Sound Events

Contact Info

Product

Resources

About