Speaker Verification Benchmark

OpenSVBench Leaderboard

Cross-scenario evaluation of speaker verification systems on a fixed public core. The global ranking uses source-balanced scenario-macro EER; lower is better.

18 ranked systems 11 scenarios 26 trial sets 16 source datasets
Fixed-Core Ranking

Global ranking

Only reviewed, completed submissions with full current-core coverage are included.

Current leader: wespeaker/w2vbert2 · Public checkpoint
Rank System Submitted by EER ↓ Scenario macro minDCF ↓ Scenario macro Stronger Scenarios Weaker Scenarios
#1 wespeaker/w2vbert2
vb2-w2vbert
Public checkpoint 9.9604 0.360767
#2 palabraai/redimnet2-b6
redimnet2
Public checkpoint 10.5277 0.419432
#3 wespeaker/samresnet100
vb2-sam100
Public checkpoint 10.5780 0.390239
#4 wespeaker/res152-voxceleb
wespeaker_resnet152
Public checkpoint 11.2275 0.442748
#5 wespeaker/res293-voxceleb
wespeaker_resnet293
Public checkpoint 11.3130 0.439607
#6 iic/eres2netv2-zh
eresv2_zh
Public checkpoint 11.5449 0.484961
#7 wespeaker/res34-voxceleb
wespeaker_resnet34_voxceleb
Public checkpoint 12.0516 0.477054
#8 iic/eres2net-en
eres_en
Public checkpoint 12.0955 0.472802
#9 wespeaker/campplus-voxceleb
campplus_wespeaker
Public checkpoint 13.4165 0.520675
#10 speechbrain/ecapa
ecapa
Public checkpoint 13.5461 0.543660
#11 wespeaker/ecapa1024-voxceleb
wespeaker_ecapa1024
Public checkpoint 14.0835 0.534111
#12 iic/eres2net-large-3dspeaker
eres2net_large_3dspeaker
Public checkpoint 14.1682 0.593516
#13 wespeaker/ecapa512-voxceleb
wespeaker_ecapa512
Public checkpoint 14.3120 0.541141
#14 wespeaker/res34-cnceleb
wespeaker_r34
Public checkpoint 14.4910 0.607964
#15 iic/campplus
campplus
Public checkpoint 15.1677 0.590178
#16 speechbrain/xvector
xvector
Public checkpoint 23.3539 0.776078
#17 microsoft/wavlm-base-plus-sv
wavlm_base
Public checkpoint 25.1577 0.875090
#18 microsoft/unispeech-sat-base-plus-sv
unispeech_sat_base_plus
Public checkpoint 25.6711 0.890736
Diagnostic Views

Scenario overview

Use scenario rankings to inspect deployment-specific strengths and failure modes hidden by the aggregate score.

Open full scenario index
Scenario Evaluation Focus Trial Sets Current Leader
Accent / Dialect Robustness
Accent/dialect
Accent and dialect variation 2 wespeaker/w2vbert2
Public checkpoint
Aging Robustness
Aging
Speaker aging and time gaps 2 wespeaker/samresnet100
Public checkpoint
Channel / Device Robustness
Channel/device
Device and channel variation 2 iic/eres2net-large-3dspeaker
Public checkpoint
Cross-Lingual Robustness
Cross-lingual
Language mismatch 1 wespeaker/w2vbert2
Public checkpoint
Distance Robustness
Distance
Distance mismatch 4 iic/eres2net-large-3dspeaker
Public checkpoint
Genre-Shift Robustness
Genre shift
Source-genre variation 1 iic/eres2netv2-zh
Public checkpoint
In-The-Wild Robustness
In-the-wild
Open-domain media speech 4 wespeaker/w2vbert2
Public checkpoint
Noise / Reverb Robustness
Noise/reverb
Noise and reverberation 1 wespeaker/w2vbert2
Public checkpoint
Overlap Robustness
Overlap
Overlapping speakers 2 wespeaker/w2vbert2
Public checkpoint
Short-Duration Robustness
Short-duration
Short-duration speech 4 wespeaker/w2vbert2
Public checkpoint
Speaking-Style Robustness
Speaking style
Speaking style shift 3 wespeaker/w2vbert2
Public checkpoint