Trial Set
GLOBE
No upstream SV trial keys are used; trials are generated from a frozen 10k-utterance subset spanning selected Hugging Face train and test shards.
- Source dataset
- GLOBE
- Trials
- 56848
- Protocol source
- Self-constructed protocol
- Source split
- GLOBE_V2 canonical HF train+test shards
Source Dataset
About GLOBE
GLOBE is a filtered and metadata-enriched English speech corpus derived from Common Voice. It was introduced for zero-shot speaker-adaptive text-to-speech, with broad accent coverage; it is not an official speaker-verification benchmark. OpenSVBench uses a frozen 10,000-utterance subset to construct its own accent-comparison trials.
- Language
- English
- Scale
- 535 hours; 23,519 speakers
- Accent coverage
- 164 accents
- Original task
- Zero-shot speaker-adaptive text-to-speech
Direct Ranking
Systems on this trial set
25 reviewed systems currently have valid results.
| Rank | System | Submitted by | Trial-Set EER ↓ | Trial-Set minDCF ↓ |
|---|---|---|---|---|
| #1 |
W2V-BERT 2.0 + MHFA (VoxCeleb2 + VoxBlink2)
Wav2Vec2-BERT
|
pengjy | 23.6455 | 0.513913 |
| #2 |
Dual-LoRA W2V-BERT2-Adapter-MFA (TidyVoice)
duallora-w2vbert2-tidyvoice-full-core
|
sgqt | 24.3034 | 0.499381 |
| #3 |
wespeaker/w2vbert2
vb2-w2vbert
|
Public checkpoint | 24.5937 | 0.512152 |
| #4 |
w2v-bert2.0 + cc + ecapa-tdnn + astp + sf2
w2v-bert2.0 + cc + ecapa-tdnn + astp + sf2
|
JinghanPeng | 24.6130 | 0.505437 |
| #5 |
wespeaker/samresnet100
vb2-sam100
|
Public checkpoint | 24.8065 | 0.531753 |
| #6 |
W2V-BERT 2.0 + Adapter-MFA (cross-device consistency FT + weight soup)
ActionPower VoicePrint
|
ActionPower | 25.0000 | 0.500793 |
| #7 |
Agora Voiceprint (ResNet293)
agora_vp
|
agora | 25.0193 | 0.505786 |
| #8 |
palabraai/redimnet2-b6
redimnet2
|
Public checkpoint | 25.0193 | 0.533495 |
| #9 |
AntSpeaker/MECT-B2
antspeaker-mect-b2-vb2-full-core
|
Public checkpoint | 25.0484 | 0.505824 |
| #10 |
wespeaker/res34-voxceleb
wespeaker_resnet34_voxceleb
|
Public checkpoint | 25.0600 | 0.531734 |
| #11 |
wespeaker/res152-voxceleb
wespeaker_resnet152
|
Public checkpoint | 25.0967 | 0.533224 |
| #12 |
iic/eres2net-en
eres_en
|
Public checkpoint | 25.1935 | 0.535991 |
| #13 |
MECT-b2-czur-finetuned
MECT-b2-czur-finetuned
|
czur | 25.1935 | 0.495685 |
| #14 |
palabraai/redimnet2-b6-vb2+vox2
redimnet2-b6-vb2vox2-lm-full-core
|
Public checkpoint | 25.2515 | 0.507624 |
| #15 |
wespeaker/ecapa512-voxceleb
wespeaker_ecapa512
|
Public checkpoint | 25.4257 | 0.525755 |
| #16 |
speechbrain/ecapa
ecapa
|
Public checkpoint | 25.4450 | 0.533282 |
| #17 |
wespeaker/res293-voxceleb
wespeaker_resnet293
|
Public checkpoint | 25.4837 | 0.531618 |
| #18 |
wespeaker/campplus-voxceleb
campplus_wespeaker
|
Public checkpoint | 25.6966 | 0.530070 |
| #19 |
wespeaker/ecapa1024-voxceleb
wespeaker_ecapa1024
|
Public checkpoint | 25.7740 | 0.541118 |
| #20 |
wespeaker/res34-cnceleb
wespeaker_r34
|
Public checkpoint | 26.3293 | 0.658746 |
| #21 |
iic/eres2netv2-zh
eresv2_zh
|
Public checkpoint | 26.7067 | 0.580650 |
| #22 |
microsoft/unispeech-sat-base-plus-sv
unispeech_sat_base_plus
|
Public checkpoint | 26.9350 | 0.824245 |
| #23 |
microsoft/wavlm-base-plus-sv
wavlm_base
|
Public checkpoint | 27.0704 | 0.791331 |
| #24 |
iic/campplus
campplus
|
Public checkpoint | 27.7090 | 0.634462 |
| #25 |
iic/eres2net-large-3dspeaker
eres2net_large_3dspeaker
|
Public checkpoint | 28.8893 | 0.801800 |