Domain Attentive Fusion for End-to-end Dialect Identification with Unknown Target Domain

Suwon Shon, Ahmed Ali, James Glass

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

End-to-end deep learning language or dialect identification systems operate on the spectrogram or other acoustic feature and directly generate identification scores for each class. An important issue for end-to-end systems is to have some knowledge of the application domain, because the system can be vulnerable to use cases that were not seen in the training phase; such a scenario is often referred to as a domain mismatched condition. In general, we assume that there is enough variation in the training dataset to expose the system to multiple domains. In this work, we study how to best make use a training dataset in order to have maximum effectiveness on unknown target domains. Our goal is to process the input without any knowledge of the target domain while preserving robust performance on other domains as well. To accomplish this objective, we propose a domain attentive fusion approach for end-to-end dialect/language identification systems. To help with experimentation, we collect a dataset from three different domains, and create experimental protocols for a domain mismatched condition. The results of our proposed approach, which were tested on a variety of broadcast and YouTube data, shows significant performance gain compared to traditional approaches, even without any prior target domain information.

Original languageEnglish
Title of host publication2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages5951-5955
Number of pages5
ISBN (Electronic)9781479981311
DOIs
Publication statusPublished - 1 May 2019
Event44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Brighton, United Kingdom
Duration: 12 May 201917 May 2019

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume2019-May
ISSN (Print)1520-6149

Conference

Conference44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019
CountryUnited Kingdom
CityBrighton
Period12/5/1917/5/19

Fingerprint

Identification (control systems)
Fusion reactions
Acoustics
Deep learning

Keywords

  • Dialect identification
  • fusion
  • language identification
  • self-attention

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Shon, S., Ali, A., & Glass, J. (2019). Domain Attentive Fusion for End-to-end Dialect Identification with Unknown Target Domain. In 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings (pp. 5951-5955). [8682533] (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2019-May). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2019.8682533

Domain Attentive Fusion for End-to-end Dialect Identification with Unknown Target Domain. / Shon, Suwon; Ali, Ahmed; Glass, James.

2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. p. 5951-5955 8682533 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2019-May).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Shon, S, Ali, A & Glass, J 2019, Domain Attentive Fusion for End-to-end Dialect Identification with Unknown Target Domain. in 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings., 8682533, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2019-May, Institute of Electrical and Electronics Engineers Inc., pp. 5951-5955, 44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019, Brighton, United Kingdom, 12/5/19. https://doi.org/10.1109/ICASSP.2019.8682533
Shon S, Ali A, Glass J. Domain Attentive Fusion for End-to-end Dialect Identification with Unknown Target Domain. In 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. p. 5951-5955. 8682533. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2019.8682533
Shon, Suwon ; Ali, Ahmed ; Glass, James. / Domain Attentive Fusion for End-to-end Dialect Identification with Unknown Target Domain. 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 5951-5955 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).
@inproceedings{90483b7c64b54efea74ef7f1c4a38dc8,
title = "Domain Attentive Fusion for End-to-end Dialect Identification with Unknown Target Domain",
abstract = "End-to-end deep learning language or dialect identification systems operate on the spectrogram or other acoustic feature and directly generate identification scores for each class. An important issue for end-to-end systems is to have some knowledge of the application domain, because the system can be vulnerable to use cases that were not seen in the training phase; such a scenario is often referred to as a domain mismatched condition. In general, we assume that there is enough variation in the training dataset to expose the system to multiple domains. In this work, we study how to best make use a training dataset in order to have maximum effectiveness on unknown target domains. Our goal is to process the input without any knowledge of the target domain while preserving robust performance on other domains as well. To accomplish this objective, we propose a domain attentive fusion approach for end-to-end dialect/language identification systems. To help with experimentation, we collect a dataset from three different domains, and create experimental protocols for a domain mismatched condition. The results of our proposed approach, which were tested on a variety of broadcast and YouTube data, shows significant performance gain compared to traditional approaches, even without any prior target domain information.",
keywords = "Dialect identification, fusion, language identification, self-attention",
author = "Suwon Shon and Ahmed Ali and James Glass",
year = "2019",
month = "5",
day = "1",
doi = "10.1109/ICASSP.2019.8682533",
language = "English",
series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "5951--5955",
booktitle = "2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings",

}

TY - GEN

T1 - Domain Attentive Fusion for End-to-end Dialect Identification with Unknown Target Domain

AU - Shon, Suwon

AU - Ali, Ahmed

AU - Glass, James

PY - 2019/5/1

Y1 - 2019/5/1

N2 - End-to-end deep learning language or dialect identification systems operate on the spectrogram or other acoustic feature and directly generate identification scores for each class. An important issue for end-to-end systems is to have some knowledge of the application domain, because the system can be vulnerable to use cases that were not seen in the training phase; such a scenario is often referred to as a domain mismatched condition. In general, we assume that there is enough variation in the training dataset to expose the system to multiple domains. In this work, we study how to best make use a training dataset in order to have maximum effectiveness on unknown target domains. Our goal is to process the input without any knowledge of the target domain while preserving robust performance on other domains as well. To accomplish this objective, we propose a domain attentive fusion approach for end-to-end dialect/language identification systems. To help with experimentation, we collect a dataset from three different domains, and create experimental protocols for a domain mismatched condition. The results of our proposed approach, which were tested on a variety of broadcast and YouTube data, shows significant performance gain compared to traditional approaches, even without any prior target domain information.

AB - End-to-end deep learning language or dialect identification systems operate on the spectrogram or other acoustic feature and directly generate identification scores for each class. An important issue for end-to-end systems is to have some knowledge of the application domain, because the system can be vulnerable to use cases that were not seen in the training phase; such a scenario is often referred to as a domain mismatched condition. In general, we assume that there is enough variation in the training dataset to expose the system to multiple domains. In this work, we study how to best make use a training dataset in order to have maximum effectiveness on unknown target domains. Our goal is to process the input without any knowledge of the target domain while preserving robust performance on other domains as well. To accomplish this objective, we propose a domain attentive fusion approach for end-to-end dialect/language identification systems. To help with experimentation, we collect a dataset from three different domains, and create experimental protocols for a domain mismatched condition. The results of our proposed approach, which were tested on a variety of broadcast and YouTube data, shows significant performance gain compared to traditional approaches, even without any prior target domain information.

KW - Dialect identification

KW - fusion

KW - language identification

KW - self-attention

UR - http://www.scopus.com/inward/record.url?scp=85068993424&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068993424&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2019.8682533

DO - 10.1109/ICASSP.2019.8682533

M3 - Conference contribution

AN - SCOPUS:85068993424

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 5951

EP - 5955

BT - 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -