Trial Set

GLOBE

Ranking for the GLOBE trial set from the GLOBE dataset.

GLOBE globe Accent/dialect

Trials 56848

Protocol source Self-constructed protocol

Source split GLOBE_V2 canonical HF train+test shards

No upstream SV trial keys are used; trials are generated from a frozen 10k-utterance subset spanning selected Hugging Face train and test shards.

Ranking

Reviewed rows

18 rows

Showing maintainer-approved full-core submissions in the public ranking.

Current Leader

24.5937

trial-set EER

0.512152 trial-set minDCF

Trial-Set Ranking

This table keeps the raw trial-set ranking visible, with no scenario-level aggregation applied.

Rank	System	Rank Signal	Trial-Set EER	Trial-Set minDCF
#1	wespeaker/w2vbert2 vb2-w2vbert	Leader	24.5937	0.512152
#2	wespeaker/samresnet100 vb2-sam100	Top 3	24.8065	0.531753
#3	palabraai/redimnet2-b6 redimnet2	Top 3	25.0193	0.533495
#4	wespeaker/res34-voxceleb wespeaker_resnet34_voxceleb	Competitive	25.0600	0.531734
#5	wespeaker/res152-voxceleb wespeaker_resnet152	Competitive	25.0967	0.533224
#6	iic/eres2net-en eres_en	Competitive	25.1935	0.535991
#7	wespeaker/ecapa512-voxceleb wespeaker_ecapa512	Competitive	25.4257	0.525755
#8	speechbrain/ecapa ecapa	Competitive	25.4450	0.533282
#9	wespeaker/res293-voxceleb wespeaker_resnet293	Competitive	25.4837	0.531618
#10	wespeaker/campplus-voxceleb campplus_wespeaker	Needs work	25.6966	0.530070
#11	wespeaker/ecapa1024-voxceleb wespeaker_ecapa1024	Needs work	25.7740	0.541118
#12	wespeaker/res34-cnceleb wespeaker_r34	Needs work	26.3293	0.658746
#13	iic/eres2netv2-zh eresv2_zh	Needs work	26.7067	0.580650
#14	microsoft/unispeech-sat-base-plus-sv unispeech_sat_base_plus	Needs work	26.9350	0.824245
#15	microsoft/wavlm-base-plus-sv wavlm_base	Needs work	27.0704	0.791331
#16	speechbrain/xvector xvector	Needs work	27.2639	0.658882
#17	iic/campplus campplus	Needs work	27.7090	0.634462
#18	iic/eres2net-large-3dspeaker eres2net_large_3dspeaker	Needs work	28.8893	0.801800