A multi-device dataset for urban acoustic scene classification

By Annamaria Mesaros, Toni Heittola, Tuomas Virtanen. 2018

434 citations recorded by the source corpus.

This paper introduces the acoustic scene classification task of DCASE 2018 Challenge and the TUT Urban Acoustic Scenes 2018 dataset provided for the task, and evaluates the performance of a baseline system in the task. As in previous years of the challenge, the task is defined for classification of short audio samples into one of predefined acoustic scene classes, using a supervised, closed-set classification setup. The newly recorded TUT Urban Acoustic Scenes 2018 dataset consists of ten different acoustic scenes and was recorded in six large European cities, therefore it has a higher acoustic variability than the previous datasets used for this task, and in addition to high-quality binaural recordings, it also includes data recorded with mobile devices. We also present the baseline system consisting of a convolutional neural network and its performance in the subtasks using the recommended cross-validation setup.

Subfields: Computer Vision and Pattern Recognition, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

FSD50K: An Open Dataset of Human-Labeled Sound Events (2020, background): “Some recent large audio datasets do not provide predeﬁned validation sets [26, 129] allowing dataset users to create their own.”
PANNs: Large-Scale Pretrained Audio Neural Networks for Audio Pattern Recognition (2019, methodology): “as audio tagging [1], acoustic scene classification [2], music classification [3], speech emotion classification and sound event”
Unsupervised adversarial domain adaptation for acoustic scene classification (2018, methodology): “For model M , we employ two different models; one that achieved the first place in an acoustic scene classification contest(1) and a second that is the published baseline model for the Task 1, subtask B, of the DCASE challenge 2018 [22].”
VOICe: A Sound Event Detection Dataset For Generalizable Domain Adaptation (2019, methodology): “As the audio samples of acoustic scenes, we use the development set of the TAU Urban Acoustic Scenes 2019 dataset [19].”
Hearing aid Research Data Set for Acoustic Environment Recognition (2020, background): “For instance, some of the scenes the DCASE challenge 2018 defines are traveling by a tram, traveling by a bus, traveling by an underground metro.”
Audio-Visual Model Distillation Using Acoustic Images (2019, methodology): “Current methods, even best deep learning models, lead to very low classification accuracies [14, 29] in such conditions.”
Training general-purpose audio tagging networks with noisy labels and iterative self-verification (2018, methodology): “Technically, the task at hand is similar to other tasks (of earlier versions) of the DCASE challenge, which focus on detection or classification of audio scenes [4].”
An Adversarial Feature Distillation Method for Audio Classification (2019, positive): “The DCASE18 task 1A [32] aims to classify audio scene of 10 predefined environment classes such as ‘‘park’’, ’’metro station’’.”