@inproceedings{5634be5f590c4d0abe000aab2d21fbe6,
title = "Intersession variability compensation for language detection",
abstract = "Gaussian mixture models (GMM) have become one of the standard acoustic approaches for Language Detection. These models are typically incorporated to produce a log-likelihood ratio (LLR) verification statistic. In this framework, the intersession variability within each language becomes an adverse factor degrading the accuracy. To address this problem, we formulate the LLR as a function of the GMM parameters concatenated into normalized mean supervectors, and estimate the distribution of each language in this (high dimensional) supervector space. The goal is to de-emphasize the directions with the largest intersession variability. We compare this method with two other popular intersession variability compensation methods known as Nuisance Attribute Projection (NAP) and Within-Class Covariance Normalization (WCCN). Experiments on the NIST LRE 2003 and NIST LRE 2005 speech corpora show that the presented technique reduces the error by 50% relative to the baseline, and performs competitively with the NAP and WCCN approaches. Fusion results with a phonotactic component are also presented.",
keywords = "ISV, NAP, WCCN-LLR",
author = "Xi Zhou and Ji{\v r}{\'i} Navr{\'a}til and Pelecanos, {Jason W.} and Ramaswamy, {Ganesh N.} and Huang, {Thomas S.}",
year = "2008",
doi = "10.1109/ICASSP.2008.4518570",
language = "English (US)",
isbn = "1424414849",
series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
pages = "4157--4160",
booktitle = "2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP",
note = "2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP ; Conference date: 31-03-2008 Through 04-04-2008",
}