Benchmark Scenario
Genre-Shift Robustness
Source-genre variation. The scenario score aggregates its trial sets using the fixed OpenSVBench reporting protocol.
18 ranked systems
1 trial set
14.2725 leading EER
Scenario Ranking
Systems on Genre-Shift Robustness
Evidence cells list trial-set rank and EER in this order: CN-Celeb Genre.
| Rank | System | Scenario EER ↓ | Scenario minDCF ↓ | Trial-Set Evidence Rank / EER |
|---|---|---|---|---|
| #1 |
iic/eres2netv2-zh
eresv2_zh
|
14.2725 | 0.527525 | |
| #2 |
wespeaker/res34-cnceleb
wespeaker_r34
|
15.6955 | 0.591628 | |
| #3 |
iic/campplus
campplus
|
16.8815 | 0.610655 | |
| #4 |
wespeaker/samresnet100
vb2-sam100
|
25.1200 | 0.674765 | |
| #5 |
palabraai/redimnet2-b6
redimnet2
|
25.4550 | 0.740052 | |
| #6 |
wespeaker/w2vbert2
vb2-w2vbert
|
25.9200 | 0.656577 | |
| #7 |
wespeaker/res152-voxceleb
wespeaker_resnet152
|
26.2472 | 0.806172 | |
| #8 |
wespeaker/res34-voxceleb
wespeaker_resnet34_voxceleb
|
26.5125 | 0.822805 | |
| #9 |
wespeaker/res293-voxceleb
wespeaker_resnet293
|
26.8575 | 0.811767 | |
| #10 |
iic/eres2net-en
eres_en
|
27.7395 | 0.802170 | |
| #11 |
iic/eres2net-large-3dspeaker
eres2net_large_3dspeaker
|
28.4850 | 0.888960 | |
| #12 |
speechbrain/ecapa
ecapa
|
29.5690 | 0.883898 | |
| #13 |
wespeaker/campplus-voxceleb
campplus_wespeaker
|
29.6497 | 0.829445 | |
| #14 |
wespeaker/ecapa512-voxceleb
wespeaker_ecapa512
|
30.8550 | 0.849922 | |
| #15 |
wespeaker/ecapa1024-voxceleb
wespeaker_ecapa1024
|
31.1025 | 0.910323 | |
| #16 |
speechbrain/xvector
xvector
|
36.7857 | 0.987563 | |
| #17 |
microsoft/unispeech-sat-base-plus-sv
unispeech_sat_base_plus
|
37.0343 | 0.985940 | |
| #18 |
microsoft/wavlm-base-plus-sv
wavlm_base
|
38.3800 | 0.975515 |