13th Annual Conference of the International Speech Communication Association

Portland, OR, USA
September 9-13, 2012

Anchor Models and WCCN Normalization For Speaker Trait Classification

Yazid Attabi (1,2), Pierre Dumouchel (1,2)

(1) École de technologie supérieure, Montréal, Canada
(2) Centre de recherche informatique de Montréal, Montréal, Canada

This paper presents an improved version of anchor model applied to solve the two-class classification tasks of the INTERSPEECH 2012 speaker trait Challenge. The introduction of within-class covariance normalization applied to the log-likelihood scores of the anchor space can not only improve the results compared to the unnormalized version but also exceed the performance of GMM or GMM-UBM systems. Furthermore, our results on development set show a relative improvement of 6.1%, 8.6% and 3.2% for the Personality, likability and pathology sub-challenges respectively compared to the best baseline systems provided by the organizers.

Index Terms: anchor model, WCCN, speaker trait classification, GMM model, Interspeech 2012 challenge

Full Paper

Bibliographic reference.  Attabi, Yazid / Dumouchel, Pierre (2012): "Anchor models and WCCN normalization for speaker trait classification", In INTERSPEECH-2012, 522-525.