Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/77467
PIRA download icon_1.1View/Download Full Text
Title: Denoised senone I-Vectors for robust speaker verification
Authors: Tan, Z 
Mak, MW 
Mak, BKW
Zhu, Y
Issue Date: Apr-2018
Source: IEEE/ACM transactions on audio, speech, and language processing, Apr. 2018, v. 26, no. 4, 8269399, p. 820-830
Abstract: Recently, it has been shown that senone i-vectors, whose posteriors are produced by senone deep neural networks (DNNs), outperform the conventional Gaussian mixture model (GMM) i-vectors in both speaker and language recognition tasks. The success of senone i-vectors relies on the capability of the DNN to incorporate phonetic information into the i-vector extraction process. In this paper, we argue that to apply senone i-vectors in noisy environments, it is important to robustify the phonetically discriminative acoustic features and senone posteriors estimated by the DNN. To this end, we propose a deep architecture formed by stacking a deep belief network on top of a denoising autoencoder (DAE). After backpropagation fine-tuning, the network, referred to as denoising autoencoder-deep neural network (DAE-DNN), facilitates the extraction of robust phonetically-discriminitive bottleneck (BN) features and senone posteriors for i-vector extraction. We refer to the resulting i-vectors as denoised BN-based senone i-vectors. Results on NIST 2012 SRE show that senone i-vectors outperform the conventional GMM i-vectors. More interestingly, the BN features are not only phonetically discriminative, results suggest that they also contain sufficient speaker information to produce BN-based senone i-vectors that outperform the conventional senone i-vectors. This work also shows that DAE training is more beneficial to BN feature extraction than senone posterior estimation.
Keywords: Deep learning
Denoising autoencoders
I-vectors
Noise robustness
Phonetically discriminative features
Senone posteriors
Speaker verification
Publisher: Institute of Electrical and Electronics Engineers
Journal: IEEE/ACM transactions on audio, speech, and language processing 
ISSN: 2329-9290
EISSN: 2329-9304
DOI: 10.1109/TASLP.2018.2796843
Rights: © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
The following publication Z. Tan, M. Mak, B. K. Mak and Y. Zhu, "Denoised Senone I-Vectors for Robust Speaker Verification," in IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 26, no. 4, pp. 820-830, April 2018 is available at https://doi.org/10.1109/TASLP.2018.2796843.
Appears in Collections:Journal/Magazine Article

Files in This Item:
File Description SizeFormat 
Tan_Denoised_Senone_I-Vectors.pdfPre-Published version1.36 MBAdobe PDFView/Open
Open Access Information
Status open access
File Version Final Accepted Manuscript
Access
View full-text via PolyU eLinks SFX Query
Show full item record

Page views

52
Citations as of Nov 20, 2022

Downloads

7
Citations as of Nov 20, 2022

SCOPUSTM   
Citations

9
Last Week
0
Last month
Citations as of Nov 24, 2022

WEB OF SCIENCETM
Citations

7
Last Week
0
Last month
Citations as of Nov 24, 2022

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.