Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

APPLYING COMPENSATION TECHNIQUES ON I-VECTORS EXTRACTED FROM SHORT-TEST UTTERANCES FOR SPEAKER VERIFICATION USING DEEP NEURAL NETWORK

Abstract: 

We propose a method to improve speaker verification performance when a test utterance is very short. In some situations with short test utterances, performance of i-vector/probabilistic linear discriminant analysis systems degrades. The proposed method transforms short-utterance feature vectors to adequate vectors using a deep neural network, which compensate for short utterances. To reduce the dimensionality of the search space, we extract several principal components from the residual vectors between every long utterance i-vector in a development set and its truncated short utterance i-vector. Then an input i-vector of the network is transformed by linear combination of these directions. In this case, network outputs correspond to weights for linear combination of principal components. We use public speech databases to evaluate the method. The experimental results on short2-10sec condition (det6, male portion) of the NIST 2008 speaker recognition evaluation corpus show that the proposed method reduces the minimum detection cost relative to the baseline system, which uses linear discriminant analysis transformed i-vectors as features.

poster.pdf

PDF icon poster.pdf (252 downloads)
up
0 users have voted:

Paper Details

Authors:
IL-Ho Yang, Hee-Soo Heo, Sung-Hyun Yoon, and Ha-Jin Yu
Submitted On:
8 March 2017 - 11:53pm
Short Link:
Type:
Poster
Event:
Presenter's Name:
IL-Ho Yang
Paper Code:
4000

Document Files

poster.pdf

(252 downloads)

Subscribe

[1] IL-Ho Yang, Hee-Soo Heo, Sung-Hyun Yoon, and Ha-Jin Yu, "APPLYING COMPENSATION TECHNIQUES ON I-VECTORS EXTRACTED FROM SHORT-TEST UTTERANCES FOR SPEAKER VERIFICATION USING DEEP NEURAL NETWORK", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1715. Accessed: Jun. 25, 2017.
@article{1715-17,
url = {http://sigport.org/1715},
author = {IL-Ho Yang; Hee-Soo Heo; Sung-Hyun Yoon; and Ha-Jin Yu },
publisher = {IEEE SigPort},
title = {APPLYING COMPENSATION TECHNIQUES ON I-VECTORS EXTRACTED FROM SHORT-TEST UTTERANCES FOR SPEAKER VERIFICATION USING DEEP NEURAL NETWORK},
year = {2017} }
TY - EJOUR
T1 - APPLYING COMPENSATION TECHNIQUES ON I-VECTORS EXTRACTED FROM SHORT-TEST UTTERANCES FOR SPEAKER VERIFICATION USING DEEP NEURAL NETWORK
AU - IL-Ho Yang; Hee-Soo Heo; Sung-Hyun Yoon; and Ha-Jin Yu
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1715
ER -
IL-Ho Yang, Hee-Soo Heo, Sung-Hyun Yoon, and Ha-Jin Yu. (2017). APPLYING COMPENSATION TECHNIQUES ON I-VECTORS EXTRACTED FROM SHORT-TEST UTTERANCES FOR SPEAKER VERIFICATION USING DEEP NEURAL NETWORK. IEEE SigPort. http://sigport.org/1715
IL-Ho Yang, Hee-Soo Heo, Sung-Hyun Yoon, and Ha-Jin Yu, 2017. APPLYING COMPENSATION TECHNIQUES ON I-VECTORS EXTRACTED FROM SHORT-TEST UTTERANCES FOR SPEAKER VERIFICATION USING DEEP NEURAL NETWORK. Available at: http://sigport.org/1715.
IL-Ho Yang, Hee-Soo Heo, Sung-Hyun Yoon, and Ha-Jin Yu. (2017). "APPLYING COMPENSATION TECHNIQUES ON I-VECTORS EXTRACTED FROM SHORT-TEST UTTERANCES FOR SPEAKER VERIFICATION USING DEEP NEURAL NETWORK." Web.
1. IL-Ho Yang, Hee-Soo Heo, Sung-Hyun Yoon, and Ha-Jin Yu. APPLYING COMPENSATION TECHNIQUES ON I-VECTORS EXTRACTED FROM SHORT-TEST UTTERANCES FOR SPEAKER VERIFICATION USING DEEP NEURAL NETWORK [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1715