Skip to main content

End-to-End Language Identification Using High-Order Utterance Representation with Bilinear Pooling

Jin, Ma, Song, Yan, McLoughlin, Ian Vince, Guo, Wu, Dai, Li-Rong (2017) End-to-End Language Identification Using High-Order Utterance Representation with Bilinear Pooling. In: The proceedings of Interspeech 2017. . pp. 2571-2575. International Speech Communication Society (doi:10.21437/Interspeech.2017-44)

PDF - Author's Accepted Manuscript
Download (989kB) Preview
[img]
Preview
Official URL
http://dx.doi.org/10.21437/Interspeech.2017-44

Abstract

A key problem in spoken language identification (LID) is how to design effective representations which are specific to language information. Recent advances in deep neural networks have led to significant improvements in results, with deep end-to-end methods proving effective. This paper proposes a novel network which aims to model an effective representation for high (first and second)-order statistics of LID-senones, defined as being LID analogues of senones in speech recognition. The high-order information extracted through bilinear pooling is robust to speakers, channels and background noise. Evaluation with NIST LRE 2009 shows improved performance compared to current state-of-the-art DBF/i-vector systems, achieving over 33% and 20% relative equal error rate (EER) improvement for 3s and 10s utterances and over 40% relative Cavg improvement for all durations.

Item Type: Conference or workshop item (Paper)
DOI/Identification number: 10.21437/Interspeech.2017-44
Subjects: T Technology > T Technology (General)
Divisions: Faculties > Sciences > School of Computing > Data Science
Depositing User: Ian McLoughlin
Date Deposited: 23 May 2017 08:32 UTC
Last Modified: 09 Jul 2019 11:18 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/61814 (The current URI for this page, for reference purposes)
McLoughlin, Ian Vince: https://orcid.org/0000-0001-7111-2008
  • Depositors only (login required):

Downloads

Downloads per month over past year