Codes
GCDS (Gaussianized cosine distance scoring) Algorithm
Download: code; demo data 2; demo data3
Reference:
If you found this code is useful, please cite the papers like the following:
Gang Liu, Taufiq Hasan, Hynek Boril, John H.L. Hansen, "An investigation on back-end for speaker recognition in multi-session enrollment", in Proc. ICASSP, Vancouver, Canada, pp. 7755-7759, 2013
Taufiq Hasan, Seyed O. Sadjadi, Gang Liu, Navid Shokouhi, Hynek Boril, John H.L. Hansen, "CRSS systems for 2012 nist speaker recognition evaluation", in Proc. ICASSP, Vancouver, Canada, pp. 6783-6787, 2013. (Best Paper Award)
Benchmark test (EER%). Gaussian Backend and PLDA1 backend is the state-of-art backend for LRE and SRE, respectively.
FAQ:
1) What is PLDA?
A concise description of this is:
Prazak, Jan, and Jan Silovsky. "Speaker diarization using PLDA-based speaker clustering." Intelligent Data Acquisition and Advanced Computing Systems (IDAACS), 2011 IEEE 6th International Conference on. Vol. 1. IEEE, 2011.
The related extraction is copied here: