From cslt Wiki
Revision as of 02:52, 22 May 2014 by Wangjun (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

weekly summary

1\prepare cohort data for speaker verification task, using all cross trials between fisher and enrollment and select the top score as imposters.

2\running bottleneck feature extractor

next week

1\to test bottleneck feature

2\to test NN-based logistic regression structure for speaker verification