Please use this identifier to cite or link to this item:
http://hdl.handle.net/10397/95586
DC Field | Value | Language |
---|---|---|
dc.contributor | Department of Electronic and Information Engineering | - |
dc.creator | Li, N | en_US |
dc.creator | Mak, MW | en_US |
dc.creator | Lin, WW | en_US |
dc.creator | Chien, JT | en_US |
dc.date.accessioned | 2022-09-22T06:13:59Z | - |
dc.date.available | 2022-09-22T06:13:59Z | - |
dc.identifier.issn | 0885-2308 | en_US |
dc.identifier.uri | http://hdl.handle.net/10397/95586 | - |
dc.language.iso | en | en_US |
dc.publisher | Academic Press | en_US |
dc.rights | © 2017 Elsevier Ltd. All rights reserved. | en_US |
dc.rights | © 2017. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/. | en_US |
dc.rights | The following publication Li, N., Mak, M. W., Lin, W. W., & Chien, J. T. (2017). Discriminative subspace modeling of SNR and duration variabilities for robust speaker verification. Computer Speech & Language, 45, 83-103 is available at https://doi.org/10.1016/j.csl.2017.04.001. | en_US |
dc.subject | Duration variation | en_US |
dc.subject | I-vector | en_US |
dc.subject | PLDA | en_US |
dc.subject | SNR mismatch | en_US |
dc.subject | Speaker verification | en_US |
dc.subject | Variational Bayes | en_US |
dc.title | Discriminative subspace modeling of SNR and duration variabilities for robust speaker verification | en_US |
dc.type | Journal/Magazine Article | en_US |
dc.identifier.spage | 83 | en_US |
dc.identifier.epage | 103 | en_US |
dc.identifier.volume | 45 | en_US |
dc.identifier.doi | 10.1016/j.csl.2017.04.001 | en_US |
dcterms.abstract | Although i-vectors together with probabilistic LDA (PLDA) have achieved a great success in speaker verification, how to suppress the undesirable effects caused by the variability in utterance length and background noise level is still a challenge. This paper aims to improve the robustness of i-vector based speaker verification systems by compensating for the utterance-length variability and noise-level variability. Inspired by the recent findings that noise-level variability can be modeled by a signal-to-noise ratio (SNR) subspace and that duration variability can be modeled as additive noise in the i-vector space, we propose to add an SNR factor and a duration factor to the PLDA model. In this framework, we assume that i-vectors derived from utterances with comparable durations share similar duration-specific information and that i-vectors extracted from utterances within a narrow SNR range have similar SNR-specific information. Based on these assumptions, an i-vector can be represented as a linear combination of four components: speaker, SNR, duration, and channel. A variational Bayes algorithm is developed to infer this latent variable model via a discriminative subspace training procedure. In the testing stage, different variabilities are compensated for when computing the likelihood ratio. Experiments on Common Conditions 1 and 4 in NIST 2012 SRE show that the proposed model outperforms the conventional PLDA and SNR-invariant PLDA. Results also show that the proposed model performs better than the uncertainty-propagation PLDA (UP-PLDA) for long test utterances. | - |
dcterms.accessRights | open access | en_US |
dcterms.bibliographicCitation | Computer speech and language, Sept. 2017, v. 45, p. 83-103 | en_US |
dcterms.isPartOf | Computer speech and language | en_US |
dcterms.issued | 2017-09 | - |
dc.identifier.scopus | 2-s2.0-85018305015 | - |
dc.description.validate | 202209_bcww | - |
dc.description.oa | Accepted Manuscript | en_US |
dc.identifier.FolderNumber | EIE-0662 | - |
dc.description.fundingSource | RGC | en_US |
dc.description.pubStatus | Published | en_US |
dc.identifier.OPUS | 6741064 | - |
dc.description.oaCategory | Green (AAM) | en_US |
Appears in Collections: | Journal/Magazine Article |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Mak_Discriminative_Subspace_Modeling.pdf | Pre-Published version | 2.34 MB | Adobe PDF | View/Open |
Page views
34
Last Week
0
0
Last month
Citations as of Oct 13, 2024
Downloads
56
Citations as of Oct 13, 2024
SCOPUSTM
Citations
7
Citations as of Oct 17, 2024
WEB OF SCIENCETM
Citations
5
Citations as of Oct 10, 2024
Google ScholarTM
Check
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.