DCASE Events

Introduction

This page collects basic information about all events organized by DCASE Community over the years. More detailed information is available on the each event websites. Community has concentrated its efforts on organizing evaluation challenges and workshops since 2013.

Challenges

The number of participating teams has increased steadily in DCASE organized challenges. The following table shows teams per task type over the years:

Task 2013 2016 2017
Acoustic Scene Classification 11 35 39
Sound Event Detection / Synthetic audio 13 10 13
Sound Event Detection / Real-life audio 7 13 13
Sound Event Detection / Weakly supervised 9
Audio Tagging 9

Workshops

DCASE organized workshops have brought now for two consecutive years researchers from universities and companies together to discuss about DCASE topics. The following table shows attendances from our workshops over the years:

Workshop Attendance (Adacemic / Companies) Papers
DCASE2017 Workshop 90 (61% / 39%) 27
DCASE2016 Workshop 68 (68% / 32%) 23

2017

DCASE2017 Challenge

The third DCASE Challenge was organized between 15th March 2017 - 31th July 2017.

The challenge was organized by Tampere University of Technology in collaboration with the Carnegie Mellon University, and INRIA, and it was an official IEEE Audio and Acoustic Signal Processing (AASP) challenge. Results of the challenge were presented at the DCASE 2017 Workshop, in which selected peer-reviewed publications on challenge submission were also presented.

Tasks

  • Task 1, Acoustic scene classification
  • Task 2, Detection of rare sound events
  • Task 3, Sound event detection in real life audio
  • Task 4, Large-scale weakly supervised sound event detection for smart cars

Results

Full results for all tasks can be found on the DCASE2017 Challenge website.

Participation statistics

Participants Submitted systems Authors Affiliations
Task 1 39 97 129
Task 2 13 33 38
Task 3 13 36 32
Task 4 9 34 25
Sum 74 200 224
Unique - - - -

Organizers

Annamaria Mesaros

Annamaria Mesaros

Coordinator of Task 1, Task 2, Task 3

Toni Heittola

Toni Heittola

Coordinator of Task 1, Task 2, Task 3

Aleksandr Diment

Aleksandr Diment

Coordinator of Task 2

Benjamin Elizalde

Benjamin Elizalde

Carnegie Mellon University

Coordinator of Task 4

Bhiksha Raj

Bhiksha Raj

Coordinator of Task 4

Emmanuel Vincent

Emmanuel Vincent

Coordinator of Task 4

DCASE2017 Workshop

The second DCASE Workshop was organized in conjunction with DCASE 2017 Challenge. The workshop took place in Munich on 16th and 17th of November and had a number of 90 participants (61% from academia and 39% from companies).

The technical program included two invited speakers: Shawn Hersey from Google Research and Josh McDermott from Massachusetts Institute of Technology, as well as oral and poster presentations of accepted papers. The videos an slides of the oral presentations from the workshop are available online. The full workshop proceedings are available here.

Photographs by Toni Heittola

2016

DCASE2016 Challenge

The second DCASE Challenge was organized between 8th February 2016 - 7th July 2016.

The challenge was organized by Tampere University of Technology in collaboration with the Centre for Digital Music from Queen Mary University of London, University of Surrey, and IRCCYN, and it was an official IEEE Audio and Acoustic Signal Processing (AASP) challenge. Results of the challenge were presented at the DCASE 2016 Workshop, in which selected peer-reviewed publications on challenge submission were also presented.

Tasks

  • Task 1, Acoustic scene classification
  • Task 2, Sound event detection in synthetic audio
  • Task 3, Sound event detection in real life audio
  • Task 4, Domestic audio tagging

Results

Challenge results have been published in the following journal paper:

Publication

A. Mesaros, T. Heittola, E. Benetos, P. Foster, M. Lagrange, T. Virtanen, and M. D. Plumbley. Detection and classification of acoustic scenes and events: outcome of the dcase 2016 challenge. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 26(2):379–393, Feb 2018.

PDF

Detection and Classification of Acoustic Scenes and Events: Outcome of the DCASE 2016 Challenge

Abstract

Public evaluation campaigns and datasets promote active development in target research areas, allowing direct comparison of algorithms. The second edition of the challenge on detection and classification of acoustic scenes and events (DCASE 2016) has offered such an opportunity for development of the state-of-the-art methods, and succeeded in drawing together a large number of participants from academic and industrial backgrounds. In this paper, we report on the tasks and outcomes of the DCASE 2016 challenge. The challenge comprised four tasks: acoustic scene classification, sound event detection in synthetic audio, sound event detection in real-life audio, and domestic audio tagging. We present each task in detail and analyze the submitted systems in terms of design and performance. We observe the emergence of deep learning as the most popular classification method, replacing the traditional approaches based on Gaussian mixture models and support vector machines. By contrast, feature representations have not changed substantially throughout the years, as mel frequency-based representations predominate in all tasks. The datasets created for and used in DCASE 2016 are publicly available and are a valuable resource for further research.

Keywords

Acoustics;Event detection;Hidden Markov models;Speech;Speech processing;Tagging;Acoustic scene classification;audio datasets;pattern recognition;sound event detection

PDF

Full results for all tasks can be found also on the DCASE2016 Challenge website.

Participation statistics

Participants Submitted systems Authors Affiliations
Task 1 35 49 115 55
Task 2 10 11 40 14
Task 3 13 17 46 21
Task 4 9 8 23 8
Sum 67 84 224 98
Unique - - 158 79

Organizers

Tuomas Virtanen

Tuomas Virtanen

Coordinator of Task 1, Task 3

Annamaria Mesaros

Annamaria Mesaros

Coordinator of Task 1, Task 3

Toni Heittola

Toni Heittola

Coordinator of Task 1, Task 3

Emmanouil Benetos

Emmanouil Benetos

Coordinator of Task 2

Mathieu Lagrange

Mathieu Lagrange

Coordinator of Task 2

Grégoire Lafay

Grégoire Lafay

Coordinator of Task 2

Mark D. Plumbley

Mark D. Plumbley

Coordinator of Task 4

Peter Foster

Peter Foster

Coordinator of Task 4

DCASE2016 Workshop

The first DCASE Workshop was organized in conjunction with DCASE 2016 Challenge. The workshop took place in Budapest on 3rd of September 2016 and had a number of 68 participants (68% from academia and 32% from companies).

The technical program included two invited speakers: Prof. Gael Richard from Telecom Paris Tech and Dr. Sacha Krstulovic from Audio Analytic, as well as oral and poster presentations of accepted papers. The presentations from the workshop are available online. The full workshop proceedings are available here.

Photographs by Toni Heittola

2013

DCASE 2013 Challenge

The first DCASE challenge campaign was organized between 31st March 2013 - 14th April 2013. The challenge was organized by the Centre for Digital Music and by IRCAM, under the auspices of the Audio and Acoustic Signal Processing (AASP) technical committee of the IEEE Signal Processing Society.

Results were presented at a special session in WASPAA 2013; participants were also invited to present a poster at a special session.

Tasks

  • Task 1, Acoustic scene classification
  • Task 2, Sound event detection, Office Live
  • Task 3, Sound event detection, Office Synthetic

Results

The outcomes of the DCASE 2013 challenge are now fully described in the following open-access journal article:

Publication

D. Stowell, D. Giannoulis, E. Benetos, M. Lagrange, and M.D. Plumbley. Detection and classification of acoustic scenes and events. Multimedia, IEEE Transactions on, 17(10):1733–1746, Oct 2015.

PDF

Detection and Classification of Acoustic Scenes and Events

Abstract

For intelligent systems to make best use of the audio modality, it is important that they can recognise not just speech and music, which have been researched as specific tasks, but also general sounds in everyday environments. To stimulate research in this field we conducted a public research challenge: the IEEE Audio and Acoustic Signal Processing Technical Committee challenge on Detection and Classification of Acoustic Scenes and Events (DCASE). In this paper we report on the state of the art in automatically classifying audio scenes, and automatically detecting and classifying audio events. We survey prior work as well as the state of the art represented by the submissions to the challenge from various research groups. We also provide detail on the organisation of the challenge, so that our experience as challenge hosts may be useful to those organising challenges in similar domains. We created new audio datasets and baseline systems for the challenge: these, as well as some submitted systems, are publicly available under open licenses, to serve as benchmark for further research in general-purpose machine listening.

Keywords

Event detection;Licenses;Microphones;Music;Speech;Speech recognition

PDF

Participation statistics

Participants
Task 1 11
Task 2 7
Task 3 3
Sum 21

Organizers

Mathieu Lagrange

Mathieu Lagrange

Mathias Rossignol

Mathias Rossignol