Conference paper

KESIRAJU Santosh, ONDEL Lucas, BURGET Lukáš, DEHAK Najim, KHUDANPUR Sanjeev and ČERNOCKÝ Jan. Topic identification of spoken documents using unsupervised acoustic unit discovery. In: Proceedings of ICASSP 2017. New Orleans: IEEE Signal Processing Society, 2017, pp. 5745-5749. ISBN 978-1-5090-4117-6.
Publication language:english
Original title:Topic identification of spoken documents using unsupervised acoustic unit discovery
Title (cs):Identifikace témat z mluvených dokumentů pomocí automatického hledání řečových jednotek
Pages:5745-5749
Proceedings:Proceedings of ICASSP 2017
Conference:42nd IEEE International Conference on Acoustics, Speech and Signal Processing
Place:New Orleans, US
Year:2017
ISBN:978-1-5090-4117-6
Publisher:IEEE Signal Processing Society
URL:http://www.fit.vutbr.cz/research/groups/speech/publi/2017/kesiraju_icassp2017_0005745.pdf [PDF]
Files: 
+Type Name Title Size Last modified
iconkesiraju_icassp2017_0005745.pdf205 KB2017-06-09 16:26:04
^ Select all
With selected:
Keywords
topic identification, acoustic unit discovery, unsupervised learning, non-parametric Bayesian models
Annotation
This article is about the topic identification of spoken documents using unsupervised acoustic unit discovery.
Abstract
This paper investigates the application of unsupervised acoustic unit discovery for topic identification (topic ID) of spoken audio documents. The acoustic unit discovery method is based on a nonparametric Bayesian phone-loop model that segments a speech utterance into phone-like categories. The discovered phone-like (acoustic) units are further fed into the conventional topic ID framework. Using multilingual bottleneck features for the acoustic unit discovery, we show that the proposed method outperforms other systems that are based on cross-lingual phoneme recognizer.
BibTeX:
@INPROCEEDINGS{
   author = {Santosh Kesiraju and Lucas Ondel and Luk{\'{a}}{\v{s}}
	Burget and Najim Dehak and Sanjeev Khudanpur and Jan
	{\v{C}}ernock{\'{y}}},
   title = {Topic identification of spoken documents using unsupervised
	acoustic unit discovery},
   pages = {5745--5749},
   booktitle = {Proceedings of ICASSP 2017},
   year = {2017},
   location = {New Orleans, US},
   publisher = {IEEE Signal Processing Society},
   ISBN = {978-1-5090-4117-6},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php?id=11470}
}

Your IPv4 address: 54.162.60.75
Switch to IPv6 connection

DNSSEC [dnssec]