Trial Set
VoxCeleb1-H
Keeps the upstream VoxCeleb1-H verification protocol.
- Source dataset
- VoxCeleb1
- Trials
- 550894
- Protocol source
- Official protocol
- Source split
- VoxCeleb1-H
Source Dataset
About VoxCeleb1
VoxCeleb is a large-scale speaker corpus collected from online interviews and other media. Its recordings retain real-world variation in microphones, channels, background sound, and speaking conditions. VoxCeleb1-O, E, and H use established VoxCeleb verification protocols; the short-duration entry on OpenSVBench is an additional benchmark protocol.
- Language
- Predominantly English
- VoxCeleb1
- 1,251 speakers and over 100,000 recordings
- Source
- Speech segments mined from online videos
- Conditions
- In-the-wild channel and background variation
Direct Ranking
Systems on this trial set
23 reviewed systems currently have valid results.
| Rank | System | Submitted by | Trial-Set EER ↓ | Trial-Set minDCF ↓ |
|---|---|---|---|---|
| #1 |
AntSpeaker/MECT-B2
antspeaker-mect-b2-vb2-full-core
|
Public checkpoint | 0.5387 | 0.031688 |
| #2 |
palabraai/redimnet2-b6-vb2+vox2
redimnet2-b6-vb2vox2-lm-full-core
|
Public checkpoint | 0.6788 | 0.038685 |
| #3 |
wespeaker/w2vbert2
vb2-w2vbert
|
Public checkpoint | 0.7298 | 0.041650 |
| #4 |
W2V-BERT 2.0 + MHFA (VoxCeleb2 + VoxBlink2)
Wav2Vec2-BERT
|
pengjy | 0.7583 | 0.048047 |
| #5 |
wespeaker/samresnet100
vb2-sam100
|
Public checkpoint | 0.8665 | 0.049074 |
| #6 |
Dual-LoRA W2V-BERT2-Adapter-MFA (TidyVoice)
duallora-w2vbert2-tidyvoice-full-core
|
shangguanqituan | 0.9380 | 0.055064 |
| #7 |
speechbrain/ecapa
ecapa
|
Public checkpoint | 0.9731 | 0.062575 |
| #8 |
palabraai/redimnet2-b6
redimnet2
|
Public checkpoint | 0.9892 | 0.061095 |
| #9 |
Agora Voiceprint (ResNet293)
agora_vp
|
agora_vp | 1.2872 | 0.073510 |
| #10 |
wespeaker/res152-voxceleb
wespeaker_resnet152
|
Public checkpoint | 1.2879 | 0.071894 |
| #11 |
wespeaker/res293-voxceleb
wespeaker_resnet293
|
Public checkpoint | 1.2890 | 0.075021 |
| #12 |
wespeaker/res34-voxceleb
wespeaker_resnet34_voxceleb
|
Public checkpoint | 1.6768 | 0.098504 |
| #13 |
wespeaker/campplus-voxceleb
campplus_wespeaker
|
Public checkpoint | 1.6832 | 0.098356 |
| #14 |
iic/eres2net-en
eres_en
|
Public checkpoint | 1.7866 | 0.104900 |
| #15 |
wespeaker/ecapa1024-voxceleb
wespeaker_ecapa1024
|
Public checkpoint | 1.8513 | 0.111398 |
| #16 |
wespeaker/ecapa512-voxceleb
wespeaker_ecapa512
|
Public checkpoint | 2.2570 | 0.132259 |
| #17 |
microsoft/wavlm-base-plus-sv
wavlm_base
|
Public checkpoint | 6.5121 | 0.395984 |
| #18 |
microsoft/unispeech-sat-base-plus-sv
unispeech_sat_base_plus
|
Public checkpoint | 7.4593 | 0.458766 |
| #19 |
iic/eres2netv2-zh
eresv2_zh
|
Public checkpoint | 8.9975 | 0.463313 |
| #20 |
wespeaker/res34-cnceleb
wespeaker_r34
|
Public checkpoint | 10.3836 | 0.526212 |
| #21 |
iic/campplus
campplus
|
Public checkpoint | 11.1228 | 0.548104 |
| #22 |
speechbrain/xvector
xvector
|
Public checkpoint | 13.8588 | 0.601061 |
| #23 |
iic/eres2net-large-3dspeaker
eres2net_large_3dspeaker
|
Public checkpoint | 17.1053 | 0.735645 |