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

facebooktwittermailshare

Pairwise Learning using Multi-lingual Bottleneck Features for Low-resource Query-by-example Spoken Term Detection

Abstract: 

We propose to use a feature representation obtained by pairwise learning in a low-resource language for query-by-example spoken term detection (QbE-STD). We assume that word pairs identified by humans are available in the low-resource target language. The word pairs are parameterized by a multi-lingual bottleneck feature (BNF) extractor that is trained using transcribed data in high-resource languages. The multi-lingual BNFs of the word pairs are used as an initial feature representation to train an autoencoder (AE). We extract features from an internal hidden layer of the pairwise trained AE to perform acoustic pattern matching for QbE-STD. Our experiments on the TIMIT and Switchboard corpora show that the pairwise learning brings 7.61% and 8.75% relative improvements in mean average precision (MAP) respectively over the initial feature representation.

up
0 users have voted:

Paper Details

Authors:
Yougen Yuan, Cheung-Chi Leung, Lei Xie, Hongjie Chen, Bin Ma, Haizhou Li
Submitted On:
7 March 2017 - 11:01pm
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Yougen Yuan
Paper Code:
1280
Document Year:
2017
Cite

Document Files

ICASSP2017_oral_ygyuan.pdf

(80 downloads)

Subscribe

[1] Yougen Yuan, Cheung-Chi Leung, Lei Xie, Hongjie Chen, Bin Ma, Haizhou Li, "Pairwise Learning using Multi-lingual Bottleneck Features for Low-resource Query-by-example Spoken Term Detection", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1598. Accessed: Oct. 18, 2017.
@article{1598-17,
url = {http://sigport.org/1598},
author = {Yougen Yuan; Cheung-Chi Leung; Lei Xie; Hongjie Chen; Bin Ma; Haizhou Li },
publisher = {IEEE SigPort},
title = {Pairwise Learning using Multi-lingual Bottleneck Features for Low-resource Query-by-example Spoken Term Detection},
year = {2017} }
TY - EJOUR
T1 - Pairwise Learning using Multi-lingual Bottleneck Features for Low-resource Query-by-example Spoken Term Detection
AU - Yougen Yuan; Cheung-Chi Leung; Lei Xie; Hongjie Chen; Bin Ma; Haizhou Li
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1598
ER -
Yougen Yuan, Cheung-Chi Leung, Lei Xie, Hongjie Chen, Bin Ma, Haizhou Li. (2017). Pairwise Learning using Multi-lingual Bottleneck Features for Low-resource Query-by-example Spoken Term Detection. IEEE SigPort. http://sigport.org/1598
Yougen Yuan, Cheung-Chi Leung, Lei Xie, Hongjie Chen, Bin Ma, Haizhou Li, 2017. Pairwise Learning using Multi-lingual Bottleneck Features for Low-resource Query-by-example Spoken Term Detection. Available at: http://sigport.org/1598.
Yougen Yuan, Cheung-Chi Leung, Lei Xie, Hongjie Chen, Bin Ma, Haizhou Li. (2017). "Pairwise Learning using Multi-lingual Bottleneck Features for Low-resource Query-by-example Spoken Term Detection." Web.
1. Yougen Yuan, Cheung-Chi Leung, Lei Xie, Hongjie Chen, Bin Ma, Haizhou Li. Pairwise Learning using Multi-lingual Bottleneck Features for Low-resource Query-by-example Spoken Term Detection [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1598