This example demonstrates a traditional diarization system using a sliding-
window iVector with PLDA scoring and agglomerative hierarchical clustering.
The recipe uses the following data for system development. This is in
addition to the NIST SRE 2000 dataset used for evaluation (see ../README.txt).
Corpus LDC Catalog No.
SRE2004 LDC2006S44
SRE2005 Train LDC2011S01
SRE2005 Test LDC2011S04
SRE2006 Train LDC2011S09
SRE2006 Test 1 LDC2011S10
SRE2006 Test 2 LDC2012S01
SRE2008 Train LDC2011S05
SRE2008 Test LDC2011S08
SWBD2 Phase 2 LDC99S79
SWBD2 Phase 3 LDC2002S06
SWBD Cellular 1 LDC2001S13
SWBD Cellular 2 LDC2004S07