Sciweavers

509 search results - page 98 / 102
» icassp 2008
Sort
View
ICASSP
2008
IEEE
15 years 4 months ago
Overlapped speech detection for improved speaker diarization in multiparty meetings
State-of-the-art speaker diarization systems for meetings are now at a point where overlapped speech contributes significantly to the errors made by the system. However, little i...
Kofi Boakye, B. Trueba-Hornero, Oriol Vinyals, Ger...
ICASSP
2008
IEEE
15 years 4 months ago
Universal switching portfolios under transaction costs
In this paper, we consider online (sequential) portfolio selection in a competitive algorithm framework under transaction costs. We construct a sequential algorithm for portfolio ...
Suleyman Serdar Kozat, Andrew C. Singer
ICASSP
2008
IEEE
15 years 4 months ago
Unsupervised learning of auditory filter banks using non-negative matrix factorisation
Non-negative matrix factorisation (NMF) is an unsupervised learning technique that decomposes a non-negative data matrix into a product of two lower rank non-negative matrices. Th...
Alexander Bertrand, Kris Demuynck, Veronique Stout...
ICASSP
2008
IEEE
15 years 4 months ago
Compressed sensing - a look beyond linear programming
Recently, significant attention in compressed sensing has been focused on Basis Pursuit, exchanging the cardinality operator with the l1-norm, which leads to a linear formulation...
Christian R. Berger, Javier Areta, Krishna R. Patt...
ICASSP
2009
IEEE
15 years 4 months ago
Comparing maximum a posteriori vector quantization and Gaussian mixture models in speaker verification
Gaussian mixture model - universal background model (GMMUBM) is a standard reference classifier in speaker verification. We have recently proposed a simplified model using vect...
Tomi Kinnunen, Juhani Saastamoinen, Ville Hautam&a...