Deep Neural Network based learning and transferring mid-level audio features for acoustic scene classification

Seongkyu Mun; Suwon Shon; Wooil Kim; David K. Han; Hanseok Ko

doi:10.1109/ICASSP.2017.7952265

Deep Neural Network based learning and transferring mid-level audio features for acoustic scene classification

Seongkyu Mun, Suwon Shon, Wooil Kim, David K. Han, Hanseok Ko

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

56 Citations (Scopus)

Abstract

Deep Neural Network (DNN) based transfer learning has been shown to be effective in Visual Object Classification (VOC) for complementing the deficit of target domain training samples by adapting classifiers that have been pre-trained for other large-scaled DataBase (DB). Although there exists an abundance of acoustic data, it can also be said that datasets of specific acoustic scenes are sparse for training Acoustic Scene Classification (ASC) models. By exploiting VOC DNN's ability of learning beyond its pre-trained environments, this paper proposes DNN based transfer learning for ASC. Effectiveness of the proposed method is demonstrated on the database of IEEE DCASE Challenge 2016 Task 1 and home surveillance environment via representative experiments. Its improved performance is verified by comparing it to prominent conventional methods.

Original language	English
Title of host publication	2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	796-800
Number of pages	5
ISBN (Electronic)	9781509041176
DOIs	https://doi.org/10.1109/ICASSP.2017.7952265
Publication status	Published - 2017 Jun 16
Event	2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - New Orleans, United States Duration: 2017 Mar 5 → 2017 Mar 9

Other

Other	2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017
Country/Territory	United States
City	New Orleans
Period	17/3/5 → 17/3/9

Keywords

acoustic scene classification
deep neural network
mid-level feature
Transfer learning

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/ICASSP.2017.7952265

Cite this

Mun, S., Shon, S., Kim, W., Han, D. K., & Ko, H. (2017). Deep Neural Network based learning and transferring mid-level audio features for acoustic scene classification. In 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings (pp. 796-800). Article 7952265 Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2017.7952265

Deep Neural Network based learning and transferring mid-level audio features for acoustic scene classification. / Mun, Seongkyu; Shon, Suwon; Kim, Wooil et al.
2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. p. 796-800 7952265.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Mun, S, Shon, S, Kim, W, Han, DK & Ko, H 2017, Deep Neural Network based learning and transferring mid-level audio features for acoustic scene classification. in 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings., 7952265, Institute of Electrical and Electronics Engineers Inc., pp. 796-800, 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017, New Orleans, United States, 17/3/5. https://doi.org/10.1109/ICASSP.2017.7952265

Mun S, Shon S, Kim W, Han DK, Ko H. Deep Neural Network based learning and transferring mid-level audio features for acoustic scene classification. In 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2017. p. 796-800. 7952265 doi: 10.1109/ICASSP.2017.7952265

@inproceedings{98efcb6a09f148eeb8b4def3b44c0afa,

title = "Deep Neural Network based learning and transferring mid-level audio features for acoustic scene classification",

abstract = "Deep Neural Network (DNN) based transfer learning has been shown to be effective in Visual Object Classification (VOC) for complementing the deficit of target domain training samples by adapting classifiers that have been pre-trained for other large-scaled DataBase (DB). Although there exists an abundance of acoustic data, it can also be said that datasets of specific acoustic scenes are sparse for training Acoustic Scene Classification (ASC) models. By exploiting VOC DNN's ability of learning beyond its pre-trained environments, this paper proposes DNN based transfer learning for ASC. Effectiveness of the proposed method is demonstrated on the database of IEEE DCASE Challenge 2016 Task 1 and home surveillance environment via representative experiments. Its improved performance is verified by comparing it to prominent conventional methods.",

keywords = "acoustic scene classification, deep neural network, mid-level feature, Transfer learning",

author = "Seongkyu Mun and Suwon Shon and Wooil Kim and Han, {David K.} and Hanseok Ko",

year = "2017",

month = jun,

day = "16",

doi = "10.1109/ICASSP.2017.7952265",

language = "English",

pages = "796--800",

booktitle = "2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 ; Conference date: 05-03-2017 Through 09-03-2017",

}

TY - GEN

T1 - Deep Neural Network based learning and transferring mid-level audio features for acoustic scene classification

AU - Mun, Seongkyu

AU - Shon, Suwon

AU - Kim, Wooil

AU - Han, David K.

AU - Ko, Hanseok

PY - 2017/6/16

Y1 - 2017/6/16

N2 - Deep Neural Network (DNN) based transfer learning has been shown to be effective in Visual Object Classification (VOC) for complementing the deficit of target domain training samples by adapting classifiers that have been pre-trained for other large-scaled DataBase (DB). Although there exists an abundance of acoustic data, it can also be said that datasets of specific acoustic scenes are sparse for training Acoustic Scene Classification (ASC) models. By exploiting VOC DNN's ability of learning beyond its pre-trained environments, this paper proposes DNN based transfer learning for ASC. Effectiveness of the proposed method is demonstrated on the database of IEEE DCASE Challenge 2016 Task 1 and home surveillance environment via representative experiments. Its improved performance is verified by comparing it to prominent conventional methods.

AB - Deep Neural Network (DNN) based transfer learning has been shown to be effective in Visual Object Classification (VOC) for complementing the deficit of target domain training samples by adapting classifiers that have been pre-trained for other large-scaled DataBase (DB). Although there exists an abundance of acoustic data, it can also be said that datasets of specific acoustic scenes are sparse for training Acoustic Scene Classification (ASC) models. By exploiting VOC DNN's ability of learning beyond its pre-trained environments, this paper proposes DNN based transfer learning for ASC. Effectiveness of the proposed method is demonstrated on the database of IEEE DCASE Challenge 2016 Task 1 and home surveillance environment via representative experiments. Its improved performance is verified by comparing it to prominent conventional methods.

KW - acoustic scene classification

KW - deep neural network

KW - mid-level feature

KW - Transfer learning

UR - http://www.scopus.com/inward/record.url?scp=85023778381&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85023778381&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2017.7952265

DO - 10.1109/ICASSP.2017.7952265

M3 - Conference contribution

AN - SCOPUS:85023778381

SP - 796

EP - 800

BT - 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017

Y2 - 5 March 2017 through 9 March 2017

ER -

Deep Neural Network based learning and transferring mid-level audio features for acoustic scene classification

Abstract

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this