The LIA-Eurecom RT'09 speaker diarization system : enhancements in speaker modelling and cluster purification
There are two approaches to speaker diarization. They are bottom-up and top-down. Our work on top-down systems show that they can deliver competitive results compared to bottom-up systems and that they are extremely computationally efﬁcient, but also that they are particularly prone to poor model initialisation and cluster impurities. In this paper we present enhancements to our state-of-the-art, top-down approach to speaker diarization that deliver improved stability across three different datasets composed of conference meetings from ﬁve standard NIST RT evaluations. We report an improved approach to speaker modelling which, despite having greater chances for cluster impurities, delivers a 35% relative improvement in DER for the MDM condition. We also describe new work to incorporate cluster puriﬁcation into a top-down sys- tem which delivers relative improvements of 44% over the baseline system without compromising computational efﬁciency.
ICASSP 2010, 35th International Conference on Acoustics, Speech, and Signal Processing, March 14-19, 2010, Dallas, Texas, USA ICASSP 2010, 35th International Conference on Acoustics, Speech, and Signal Processing, March 14-19, 2010, Dallas, Texas, USAproceeding with peer review 2010-03-14