8th European Conference on Speech Communication and Technology

Geneva, Switzerland
September 1-4, 2003


Novel Approaches for One- and Two-Speaker Detection

Sachin S. Kajarekar (1), Andre G. Adami (2), Hynek Hermansky (2)

(1) SRI International, USA
(2) Oregon Health & Science University, USA

The paper reviews OGI submission for NIST 2002 speaker recognition evaluation. It describes the systems submitted for one- and two-speaker detection tasks and the post-evaluation improvements. In one-speaker detection system, we present a new design of a data-driven temporal filter. We show that using few broad phonetic categories improves the performance of speaker recognition system. In post evaluation experiments, we show that combinations with complementary features and modeling techniques significantly improve the performance of the GMM-based system. In two-speaker detection system, we present a structured approach to detect speaker in the conversations.

Full Paper

Bibliographic reference.  Kajarekar, Sachin S. / Adami, Andre G. / Hermansky, Hynek (2003): "Novel approaches for one- and two-speaker detection", In EUROSPEECH-2003, 2661-2664.