ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Automatic speech segmentation using probabilistic latent component modeling

Ghosh, Sayan and Sreenivas, TV (2012) Automatic speech segmentation using probabilistic latent component modeling. In: 13th Annual Conference of the International Speech Communication Association, September 9-13, 2012, Portland, Oregon, pp. 2259-2262.

Full text not available from this repository. (Request a copy)
Official URL: http://interspeech2012.org/accepted-abstract.html?...

Abstract

Latent variable methods, such as PLCA (Probabilistic Latent Component Analysis) have been successfully used for analysis of non-negative signal representations. In this paper, we formulate PLCS (Probabilistic Latent Component Segmentation), which models each time frame of a spectrogram as a spectral distribution. Given the signal spectrogram, the segmentation boundaries are estimated using a maximum-likelihood approach. For an efficient solution, the algorithm imposes a hard constraint that each segment is modelled by a single latent component. The hard constraint facilitates the solution of ML boundary estimation using dynamic programming. The PLCS framework does not impose a parametric assumption unlike earlier ML segmentation techniques. PLCS can be naturally extended to model coarticulation between successive phones. Experiments on the TIMIT corpus show that the proposed technique is promising compared to most state of the art speech segmentation algorithms.

Item Type: Conference Paper
Publisher: ISCA - International Speech Communication Association
Additional Information: Copyright of this article belongs to ISCA - International Speech Communication Association.
Keywords: Speech Segmentation; PLCA; Spectrograms; Coarticulation; Dynamic Programming
Department/Centre: Division of Electrical Sciences > Electrical Communication Engineering
Date Deposited: 22 Sep 2013 07:07
Last Modified: 22 Sep 2013 07:07
URI: http://eprints.iisc.ac.in/id/eprint/47278

Actions (login required)

View Item View Item