DCASE 2017 challenge setup: tasks, datasets and baseline system

Annamaria Mesaros; Toni Heittola; Aleksandr Diment; Benjamin Martinez Elizalde; Ankit Shah; Emmanuel Vincent; Bhiksha Raj; Tuomas Virtanen

DCASE 2017 challenge setup: tasks, datasets and baseline system

Annamaria Mesaros
, Toni Heittola
, Aleksandr Diment
, Benjamin Martinez Elizalde
, Ankit Shah
, Emmanuel Vincent
, Bhiksha Raj
, Tuomas Virtanen

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

Abstract

DCASE 2017 Challenge consists of four tasks: acoustic scene classification, detection of rare sound events, sound event detection in real-life audio, and large-scale weakly supervised sound event detection for smart cars. This paper presents the setup of these tasks: task definition, dataset, experimental setup, and baseline system results on the development dataset. The baseline systems for all tasks rely on the same implementation using multilayer perceptron and log mel-energies, but differ in the structure of the output layer and the decision making process, as well as the evaluation of system output using task specific metrics.

Original language	English
Title of host publication	Proceedings of the Detection and Classification of Acoustic Scenes and Events 2017 Workshop (DCASE2017)
Publisher	Tampere University of Technology. Laboratory of Signal Processing
Pages	85-92
ISBN (Electronic)	978-952-15-4042-4
Publication status	Published - 2017
Publication type	A4 Article in conference proceedings
Event	Detection and Classification of Acoustic Scenes and Events Workshop - Duration: 1 Jan 2000 → …

Conference

Conference	Detection and Classification of Acoustic Scenes and Events Workshop
Period	1/01/00 → …

Keywords

Sound scene analysis
Acoustic scene classification
Sound event detection
Audio tagging
Rare sound events

Publication forum classification

No publication forum level

Access to Document

http://urn.fi/URN:ISBN:978-952-15-4042-4

Cite this

Mesaros, A., Heittola, T., Diment, A., Elizalde, B. M., Shah, A., Vincent, E., Raj, B., & Virtanen, T. (2017). DCASE 2017 challenge setup: tasks, datasets and baseline system. In Proceedings of the Detection and Classification of Acoustic Scenes and Events 2017 Workshop (DCASE2017) (pp. 85-92). Tampere University of Technology. Laboratory of Signal Processing. http://urn.fi/URN:ISBN:978-952-15-4042-4

@inproceedings{32ee7d149a974ac6a641e79bc5106c1f,

title = "DCASE 2017 challenge setup: tasks, datasets and baseline system",

abstract = "DCASE 2017 Challenge consists of four tasks: acoustic scene classification, detection of rare sound events, sound event detection in real-life audio, and large-scale weakly supervised sound event detection for smart cars. This paper presents the setup of these tasks: task definition, dataset, experimental setup, and baseline system results on the development dataset. The baseline systems for all tasks rely on the same implementation using multilayer perceptron and log mel-energies, but differ in the structure of the output layer and the decision making process, as well as the evaluation of system output using task specific metrics.",

keywords = "Sound scene analysis, Acoustic scene classification, Sound event detection, Audio tagging, Rare sound events",

author = "Annamaria Mesaros and Toni Heittola and Aleksandr Diment and Elizalde, \{Benjamin Martinez\} and Ankit Shah and Emmanuel Vincent and Bhiksha Raj and Tuomas Virtanen",

year = "2017",

language = "English",

pages = "85--92",

booktitle = "Proceedings of the Detection and Classification of Acoustic Scenes and Events 2017 Workshop (DCASE2017)",

publisher = "Tampere University of Technology. Laboratory of Signal Processing",

note = "Detection and Classification of Acoustic Scenes and Events Workshop ; Conference date: 01-01-2000",

}

Mesaros, A, Heittola, T, Diment, A, Elizalde, BM, Shah, A, Vincent, E, Raj, B & Virtanen, T 2017, DCASE 2017 challenge setup: tasks, datasets and baseline system. in Proceedings of the Detection and Classification of Acoustic Scenes and Events 2017 Workshop (DCASE2017). Tampere University of Technology. Laboratory of Signal Processing, pp. 85-92, Detection and Classification of Acoustic Scenes and Events Workshop, 1/01/00. <http://urn.fi/URN:ISBN:978-952-15-4042-4>

DCASE 2017 challenge setup: tasks, datasets and baseline system. / Mesaros, Annamaria; Heittola, Toni; Diment, Aleksandr et al.
Proceedings of the Detection and Classification of Acoustic Scenes and Events 2017 Workshop (DCASE2017). Tampere University of Technology. Laboratory of Signal Processing, 2017. p. 85-92.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

TY - GEN

T1 - DCASE 2017 challenge setup: tasks, datasets and baseline system

AU - Mesaros, Annamaria

AU - Heittola, Toni

AU - Diment, Aleksandr

AU - Elizalde, Benjamin Martinez

AU - Shah, Ankit

AU - Vincent, Emmanuel

AU - Raj, Bhiksha

AU - Virtanen, Tuomas

PY - 2017

Y1 - 2017

N2 - DCASE 2017 Challenge consists of four tasks: acoustic scene classification, detection of rare sound events, sound event detection in real-life audio, and large-scale weakly supervised sound event detection for smart cars. This paper presents the setup of these tasks: task definition, dataset, experimental setup, and baseline system results on the development dataset. The baseline systems for all tasks rely on the same implementation using multilayer perceptron and log mel-energies, but differ in the structure of the output layer and the decision making process, as well as the evaluation of system output using task specific metrics.

AB - DCASE 2017 Challenge consists of four tasks: acoustic scene classification, detection of rare sound events, sound event detection in real-life audio, and large-scale weakly supervised sound event detection for smart cars. This paper presents the setup of these tasks: task definition, dataset, experimental setup, and baseline system results on the development dataset. The baseline systems for all tasks rely on the same implementation using multilayer perceptron and log mel-energies, but differ in the structure of the output layer and the decision making process, as well as the evaluation of system output using task specific metrics.

KW - Sound scene analysis

KW - Acoustic scene classification

KW - Sound event detection

KW - Audio tagging

KW - Rare sound events

M3 - Conference contribution

SP - 85

EP - 92

BT - Proceedings of the Detection and Classification of Acoustic Scenes and Events 2017 Workshop (DCASE2017)

PB - Tampere University of Technology. Laboratory of Signal Processing

T2 - Detection and Classification of Acoustic Scenes and Events Workshop

Y2 - 1 January 2000

ER -

DCASE 2017 challenge setup: tasks, datasets and baseline system

Abstract

Conference

Keywords

Publication forum classification

Access to Document

Fingerprint

Cite this