Acoustic Scene Classification: A Competition Review

Shayan Gharib; Honain Derrar; Daisuke Niizumi; Tuukka Senttula; Janne Tommola; Toni Heittola; Tuomas Virtanen; Heikki Huttunen

doi:10.1109/MLSP.2018.8517000

Acoustic Scene Classification: A Competition Review

Shayan Gharib, Honain Derrar, Daisuke Niizumi, Tuukka Senttula, Janne Tommola, Toni Heittola, Tuomas Virtanen, Heikki Huttunen

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

9 Citations (Scopus)

Abstract

In this paper we study the problem of acoustic scene classification, i.e., categorization of audio sequences into mutually exclusive classes based on their spectral content. We describe the methods and results discovered during a competition organized in the context of a graduate machine learning course; both by the students and external participants. We identify the most suitable methods and study the impact of each by performing an ablation study of the mixture of approaches. We also compare the results with a neural network baseline, and show the improvement over that. Finally, we discuss the impact of using a competition as a part of a university course, and justify its importance in the curriculum based on student feedback.

Original language	English
Title of host publication	2018 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2018
Publisher	IEEE
ISBN (Electronic)	9781538654774
DOIs	https://doi.org/10.1109/MLSP.2018.8517000
Publication status	Published - Sept 2018
Publication type	A4 Article in conference proceedings
Event	IEEE International Workshop on Machine Learning for Signal Processing - Duration: 17 Sept 2018 → 20 Sept 2018

Conference

Conference	IEEE International Workshop on Machine Learning for Signal Processing
Period	17/09/18 → 20/09/18

Publication forum classification

Publication forum level 1

Access to Document

10.1109/MLSP.2018.8517000

https://arxiv.org/abs/1808.02357

Cite this

@inproceedings{4c79f556afe9436a973fa5766ca57c90,

title = "Acoustic Scene Classification: A Competition Review",

abstract = "In this paper we study the problem of acoustic scene classification, i.e., categorization of audio sequences into mutually exclusive classes based on their spectral content. We describe the methods and results discovered during a competition organized in the context of a graduate machine learning course; both by the students and external participants. We identify the most suitable methods and study the impact of each by performing an ablation study of the mixture of approaches. We also compare the results with a neural network baseline, and show the improvement over that. Finally, we discuss the impact of using a competition as a part of a university course, and justify its importance in the curriculum based on student feedback.",

author = "Shayan Gharib and Honain Derrar and Daisuke Niizumi and Tuukka Senttula and Janne Tommola and Toni Heittola and Tuomas Virtanen and Heikki Huttunen",

note = "INT=sgn,{"}Gharib, Shayan{"} INT=sgn,{"}Derrar, Honain{"} INT=sgn,{"}Senttula, Tuukka{"} INT=sgn,{"}Tommola, Janne{"}; IEEE International Workshop on Machine Learning for Signal Processing ; Conference date: 17-09-2018 Through 20-09-2018",

year = "2018",

month = sep,

doi = "10.1109/MLSP.2018.8517000",

language = "English",

booktitle = "2018 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2018",

publisher = "IEEE",

address = "United States",

}

Gharib, S, Derrar, H, Niizumi, D, Senttula, T, Tommola, J, Heittola, T , Virtanen, T & Huttunen, H 2018, Acoustic Scene Classification: A Competition Review. in 2018 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2018. IEEE, IEEE International Workshop on Machine Learning for Signal Processing, 17/09/18. https://doi.org/10.1109/MLSP.2018.8517000

TY - GEN

T1 - Acoustic Scene Classification

T2 - IEEE International Workshop on Machine Learning for Signal Processing

AU - Gharib, Shayan

AU - Derrar, Honain

AU - Niizumi, Daisuke

AU - Senttula, Tuukka

AU - Tommola, Janne

AU - Heittola, Toni

AU - Virtanen, Tuomas

AU - Huttunen, Heikki

N1 - INT=sgn,"Gharib, Shayan" INT=sgn,"Derrar, Honain" INT=sgn,"Senttula, Tuukka" INT=sgn,"Tommola, Janne"

PY - 2018/9

Y1 - 2018/9

N2 - In this paper we study the problem of acoustic scene classification, i.e., categorization of audio sequences into mutually exclusive classes based on their spectral content. We describe the methods and results discovered during a competition organized in the context of a graduate machine learning course; both by the students and external participants. We identify the most suitable methods and study the impact of each by performing an ablation study of the mixture of approaches. We also compare the results with a neural network baseline, and show the improvement over that. Finally, we discuss the impact of using a competition as a part of a university course, and justify its importance in the curriculum based on student feedback.

AB - In this paper we study the problem of acoustic scene classification, i.e., categorization of audio sequences into mutually exclusive classes based on their spectral content. We describe the methods and results discovered during a competition organized in the context of a graduate machine learning course; both by the students and external participants. We identify the most suitable methods and study the impact of each by performing an ablation study of the mixture of approaches. We also compare the results with a neural network baseline, and show the improvement over that. Finally, we discuss the impact of using a competition as a part of a university course, and justify its importance in the curriculum based on student feedback.

U2 - 10.1109/MLSP.2018.8517000

DO - 10.1109/MLSP.2018.8517000

M3 - Conference contribution

BT - 2018 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2018

PB - IEEE

Y2 - 17 September 2018 through 20 September 2018

ER -

Acoustic Scene Classification: A Competition Review

Abstract

Conference

Publication forum classification

Access to Document

Fingerprint

Cite this