A Compact Model for Speaker-Adaptive Training

DMCA

A Compact Model for Speaker-Adaptive Training (1996)

Cached

Download Links

[www.asel.udel.edu]

by Tasos Anastasakos , John Mcdonough , Richard Schwartz , John Makhoul

Venue:	in Proc. ICSLP
Citations:	208 - 20 self

BibTeX

@INPROCEEDINGS{Anastasakos96acompact,
    author = {Tasos Anastasakos and John Mcdonough and Richard Schwartz and John Makhoul},
    title = {A Compact Model for Speaker-Adaptive Training},
    booktitle = {in Proc. ICSLP},
    year = {1996},
    pages = {1137--1140}
}

OpenURL

Abstract

In this work we formulate a novel approach to estimating the parameters of continuous density HMMs for speaker-independent (SI) continuous speech recognition. It is motivated by the fact that variability in SI acoustic models is attributed to both phonetic variation and variation among the speakers of the training population, that is independent of the information content of the speech signal. These two variation sources are decoupled and the proposed method jointly annihilates the inter-speaker variation and estimates the HMM parameters of the SI acoustic models.

Keyphrases

compact model speaker-adaptive training si acoustic model information content phonetic variation variation source inter-speaker variation novel approach continuous speech recognition training population hmm parameter speech signal continuous density hmms

DMCA