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

facebooktwittermailshare

Exploiting Language-Mismatched Phoneme Recognizers for Unsupervised Acoustic Modeling

Abstract: 

This paper describes an investigation on acoustic modeling in the absence of transcribed training data. We propose to use language-mismatched phoneme recognizers to assist unsupervised segmentation and segment clustering of a new language. Using a language-mismatched recognizer, an input utterance is divided into many variable-length segments. Each segment is represented by a feature vector that is derived from the phoneme posterior probabilities. A spectral clustering algorithm is developed to group the segments into a prescribed number of clusters, which represent a set of basic speech units in the target language. By exploiting multiple recognizers for different languages, a wider phonetic space can be covered, leading to improved performance of segmentation and clustering. Experimental results on a multilingual speech database confirm the effectiveness of the proposed method.

up
0 users have voted:

Paper Details

Authors:
Siyuan Feng, Tan Lee, Haipeng Wang
Submitted On:
13 October 2016 - 8:27am
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Siyuan Feng
Paper Code:
118
Document Year:
2016
Cite

Document Files

slides.pdf

(885)

Subscribe

[1] Siyuan Feng, Tan Lee, Haipeng Wang, "Exploiting Language-Mismatched Phoneme Recognizers for Unsupervised Acoustic Modeling", IEEE SigPort, 2016. [Online]. Available: http://sigport.org/1170. Accessed: Feb. 18, 2020.
@article{1170-16,
url = {http://sigport.org/1170},
author = {Siyuan Feng; Tan Lee; Haipeng Wang },
publisher = {IEEE SigPort},
title = {Exploiting Language-Mismatched Phoneme Recognizers for Unsupervised Acoustic Modeling},
year = {2016} }
TY - EJOUR
T1 - Exploiting Language-Mismatched Phoneme Recognizers for Unsupervised Acoustic Modeling
AU - Siyuan Feng; Tan Lee; Haipeng Wang
PY - 2016
PB - IEEE SigPort
UR - http://sigport.org/1170
ER -
Siyuan Feng, Tan Lee, Haipeng Wang. (2016). Exploiting Language-Mismatched Phoneme Recognizers for Unsupervised Acoustic Modeling. IEEE SigPort. http://sigport.org/1170
Siyuan Feng, Tan Lee, Haipeng Wang, 2016. Exploiting Language-Mismatched Phoneme Recognizers for Unsupervised Acoustic Modeling. Available at: http://sigport.org/1170.
Siyuan Feng, Tan Lee, Haipeng Wang. (2016). "Exploiting Language-Mismatched Phoneme Recognizers for Unsupervised Acoustic Modeling." Web.
1. Siyuan Feng, Tan Lee, Haipeng Wang. Exploiting Language-Mismatched Phoneme Recognizers for Unsupervised Acoustic Modeling [Internet]. IEEE SigPort; 2016. Available from : http://sigport.org/1170