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

facebooktwittermailshare

Audio Word Similarity for Clustering with Zero Resources based on iterative HMM Classification

Abstract: 

Recent work on zero resource word discovery makes intensive use of audio fragment clustering to find repeating speech patterns. In the absence of acoustic models, the clustering step traditionally relies on dynamic time warping (DTW) to compare two samples and thus suffers from the known limitations of this technique. We propose a new sample comparison method, called 'similarity by terative classification', that exploits the modeling capacities of hidden Markov models (HMM) with no supervision.

The core idea relies on the use of HMMs trained on randomly labeled data and exploits the fact that similar samples are more likely to be classified together by a large number of random classifiers than dissimilar ones. The resulting similarity measure is compared to DTW on two tasks, namely nearest neighbor retrieval and clustering, showing that the generalization capabilities of probabilistic machine learning significantly benefit to audio word comparison and overcome many of the limitations of DTW-based comparison.

up
0 users have voted:

Paper Details

Authors:
Amélie Royer, Guillaume Gravier, Vincent Claveau
Submitted On:
19 March 2016 - 2:37pm
Short Link:
Type:
Poster
Event:
Presenter's Name:
Amélie Royer
Document Year:
2016
Cite

Document Files

Poster_ICASSP.pdf

(1602)

Subscribe

[1] Amélie Royer, Guillaume Gravier, Vincent Claveau, "Audio Word Similarity for Clustering with Zero Resources based on iterative HMM Classification", IEEE SigPort, 2016. [Online]. Available: http://sigport.org/829. Accessed: Apr. 18, 2019.
@article{829-16,
url = {http://sigport.org/829},
author = {Amélie Royer; Guillaume Gravier; Vincent Claveau },
publisher = {IEEE SigPort},
title = {Audio Word Similarity for Clustering with Zero Resources based on iterative HMM Classification},
year = {2016} }
TY - EJOUR
T1 - Audio Word Similarity for Clustering with Zero Resources based on iterative HMM Classification
AU - Amélie Royer; Guillaume Gravier; Vincent Claveau
PY - 2016
PB - IEEE SigPort
UR - http://sigport.org/829
ER -
Amélie Royer, Guillaume Gravier, Vincent Claveau. (2016). Audio Word Similarity for Clustering with Zero Resources based on iterative HMM Classification. IEEE SigPort. http://sigport.org/829
Amélie Royer, Guillaume Gravier, Vincent Claveau, 2016. Audio Word Similarity for Clustering with Zero Resources based on iterative HMM Classification. Available at: http://sigport.org/829.
Amélie Royer, Guillaume Gravier, Vincent Claveau. (2016). "Audio Word Similarity for Clustering with Zero Resources based on iterative HMM Classification." Web.
1. Amélie Royer, Guillaume Gravier, Vincent Claveau. Audio Word Similarity for Clustering with Zero Resources based on iterative HMM Classification [Internet]. IEEE SigPort; 2016. Available from : http://sigport.org/829