System Detail
wespeaker/ecapa512-voxceleb
wespeaker_ecapa512
Submitted by: Public checkpoint
- Global position
- #19 / 24
- Scenario-macro EER
- 14.3120
- Scenario-macro minDCF
- 0.541141
- Trial-set coverage
- 26 / 26
- Scenario top-three
- 0
Publicly ranked
This result is included in the reviewed public ranking.
Model Provenance
Source and architecture
- Displayed name:
wespeaker/ecapa512-voxceleb - Submitted by: Public checkpoint
- Metadata: OpenSVBench model record
- Source: WeSpeaker
- Architecture: ECAPA-TDNN
- Model family: ECAPA
- Submitted alias:
wespeaker-voxceleb-ecapa-tdnn512-LM - Created at:
2026-05-29T14:47:10.548863+00:00
Training And Links
Training data and references
- Training data: VoxCeleb2 dev, 5,994 speakers
- Training setup: ECAPA-TDNN GLOB c512 with ASTP pooling, 192-d embedding, ArcMargin, large-margin fine-tuning, speed perturbation, and MUSAN/RIRS augmentation.
- Submission mode:
full-core - Model source: Open link
- Paper: ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification
Higher-Ranked Scenarios
Scenario ranks above the global position
Show 3 supporting trial sets
- Lombard Grid Lombard: #13 on its trial-set ranking.
- GLOBE: #13 on its trial-set ranking.
- ESD: #14 on its trial-set ranking.
Lower-Ranked Scenarios
Scenario ranks below the global position
Show 3 supporting trial sets
- 3D-Speaker Distance: #21 on its trial-set ranking.
- Whisper40 Whisper: #21 on its trial-set ranking.
- CN-Celeb: #21 on its trial-set ranking.
Scenario Rankings
Full ranking across scenarios
This table shows where the model sits on each scenario, using source-balanced scenario scores and the linked trial sets as evidence.
| Scenario | Rank | Lens | Evidence | Score |
|---|---|---|---|---|
|
Aging Robustness
How stable identity representations remain across age-derived and longitudinal recording gaps.
|
#17/24
Needs work
|
Speaker aging and time gaps
3.0335 EER from leader
|
4.8252
0.212812 minDCF
|
|
|
Cross-Lingual Robustness
How well speaker identity survives enrollment-test language mismatch.
|
#17/24
Needs work
|
Language mismatch
4.8506 EER from leader
|
6.5083
0.405096 minDCF
|
|
|
Noise / Reverb Robustness
How reliably the model preserves identity when room acoustics, distractor noise, and microphone placement deviate from an easier in-corpus reference condition.
|
#17/24
Needs work
|
Noise and reverberation
5.3560 EER from leader
|
7.6160
0.264660 minDCF
|
|
|
Speaking-Style Robustness
Whether the model can preserve identity across emotion-driven change, whispered speech, and noise-induced Lombard speaking style.
|
#18/24
Needs work
|
Speaking style shift
5.1962 EER from leader
|
7.1128
0.367138 minDCF
|
|
|
Short-Duration Robustness
How much performance holds up when speech evidence is limited by duration.
|
#18/24
Needs work
|
Short-duration speech
8.6798 EER from leader
|
16.5541
0.604686 minDCF
|
|
|
Overlap Robustness
How well speaker identity survives light, mid, and heavy overlap in both meeting and domestic recordings.
|
#19/24
Needs work
|
Overlapping speakers
6.8396 EER from leader
|
14.9489
0.650355 minDCF
|
|
|
Distance Robustness
How much performance changes across meeting and domestic distance mismatch.
|
#19/24
Needs work
|
Distance mismatch
11.2542 EER from leader
|
16.0447
0.664732 minDCF
|
|
|
Accent / Dialect Robustness
How reliably the model tracks identity across accent and dialect mismatch.
|
#19/24
Needs work
|
Accent and dialect variation
9.2840 EER from leader
|
25.3328
0.762541 minDCF
|
|
|
In-The-Wild Robustness
How strong the model is on unconstrained celebrity and media speech across official CN-Celeb and VoxCeleb protocols.
|
#20/24
Needs work
|
Open-domain media speech
6.4463 EER from leader
|
8.9203
0.324306 minDCF
|
|
|
Channel / Device Robustness
How resilient the model is to device and channel mismatch.
|
#20/24
Needs work
|
Device and channel variation
14.7511 EER from leader
|
18.7142
0.846298 minDCF
|
|
|
Genre-Shift Robustness
Whether performance holds when CN-Celeb enrollment and test speech come from different source genres.
|
#20/24
Needs work
|
Source-genre variation
17.4838 EER from leader
|
30.8550
0.849922 minDCF
|
Variant Compare
Compare ECAPA variants
Expand this section to compare models in the same model family by architecture, source, training data, training setup, and leaderboard result.
Show
Compare ECAPA variants
Expand this section to compare models in the same model family by architecture, source, training data, training setup, and leaderboard result.
| Variant | Submitted by | Source | Training Data | Training Setup | Global Rank | Macro EER | Macro minDCF | Open |
|---|---|---|---|---|---|---|---|---|
|
speechbrain/ecapa
ECAPA-TDNN
|
Public checkpoint | SpeechBrain | VoxCeleb1 + VoxCeleb2 training data | SpeechBrain ECAPA-TDNN release with attentive statistical pooling and Additive Margin Softmax loss. |
#16
ranked
|
13.5461 | 0.543660 | Open |
|
wespeaker/ecapa1024-voxceleb
ECAPA-TDNN
|
Public checkpoint | WeSpeaker | VoxCeleb2 dev, 5,994 speakers | ECAPA-TDNN GLOB c1024 with ASTP pooling, 192-d embedding, ArcMargin, 150-epoch training, speed perturbation, and MUSAN/RIRS augmentation. |
#17
ranked
|
14.0835 | 0.534111 | Open |
|
wespeaker/ecapa512-voxceleb
ECAPA-TDNN
Current
|
Public checkpoint | WeSpeaker | VoxCeleb2 dev, 5,994 speakers | ECAPA-TDNN GLOB c512 with ASTP pooling, 192-d embedding, ArcMargin, large-margin fine-tuning, speed perturbation, and MUSAN/RIRS augmentation. |
#19
ranked
|
14.3120 | 0.541141 | Open |
Trial-Set Results
Raw ranking by trial set
Expand this section to inspect the detailed trial-set rankings.
Show
Raw ranking by trial set
Expand this section to inspect the detailed trial-set rankings.
| Trial Set | Rank | Scenarios | Trials | Score |
|---|---|---|---|---|
|
TidyVoiceX2-ASV
TidyVoiceX2-ASV
|
#17/24
|
Cross-lingual
|
200000 |
6.5083
0.405096 minDCF
|
|
HI-MIA
HI-MIA
|
#19/24
|
Short-duration
|
660000 |
8.5340
0.404212 minDCF
|
|
GSC Short
Google Speech Commands
|
#15/24
|
Short-duration
|
220000 |
12.8770
0.644885 minDCF
|
|
GLOBE
GLOBE
|
#13/24
|
Accent/dialect
|
56848 |
25.4257
0.525755 minDCF
|
|
CN-Celeb
CN-Celeb
|
#21/24
|
In-the-wild
|
3484292 |
16.3391
0.550287 minDCF
|
|
CN-Celeb Genre
CN-Celeb
|
#20/24
|
Genre shift
|
440000 |
30.8550
0.849922 minDCF
|
|
CN-Celeb Short
CN-Celeb
|
#20/24
|
Short-duration
|
546964 |
29.6374
0.854863 minDCF
|
|
VoxCeleb1-O
VoxCeleb1
|
#17/24
|
In-the-wild
|
37611 |
1.0371
0.085403 minDCF
|
|
VoxCeleb1-E
VoxCeleb1
|
#17/24
|
In-the-wild
|
579818 |
1.2104
0.077313 minDCF
|
|
VoxCeleb1-H
VoxCeleb1
|
#17/24
|
In-the-wild
|
550894 |
2.2570
0.132259 minDCF
|
|
VoxCeleb Short
VoxCeleb1
|
#16/24
|
Short-duration
|
394724 |
15.1680
0.514784 minDCF
|
|
3D-Speaker Device
3D-Speaker
|
#20/24
|
Channel/device
|
180000 |
25.3867
0.995540 minDCF
|
|
3D-Speaker Distance
3D-Speaker
|
#21/24
|
Distance
|
175163 |
25.4467
0.989464 minDCF
|
|
3D-Speaker Dialect
3D-Speaker
|
#19/24
|
Accent/dialect
|
180000 |
25.2400
0.999327 minDCF
|
|
FFSVC 2022 Cross-Channel
FFSVC 2022
|
#18/24
|
Channel/device
|
72000 |
12.0417
0.697056 minDCF
|
|
FFSVC 2022 Cross-Domain
FFSVC 2022
|
#18/24
|
Distance
|
66546 |
12.1783
0.703849 minDCF
|
|
Whisper40 Whisper
Whisper40
|
#21/24
|
Speaking style
|
17600 |
16.6250
0.780062 minDCF
|
|
Lombard Grid Lombard
Lombard Grid
|
#13/24
|
Speaking style
|
29524 |
0.2235
0.015499 minDCF
|
|
VOiCES Noise/Reverb
VOiCES
|
#17/24
|
Noise/reverb
|
55000 |
7.6160
0.264660 minDCF
|
|
CHiME-6 Domestic Far-Field
CHiME-6
|
#17/24
|
Distance
|
36487 |
15.1040
0.585680 minDCF
|
|
CHiME-6 Overlap
CHiME-6
|
#17/24
|
Overlap
|
39600 |
18.2778
0.716417 minDCF
|
|
ESD
ESD
|
#14/24
|
Speaking style
|
437408 |
4.4900
0.305853 minDCF
|
|
AliMeeting Near/Far
AliMeeting
|
#21/24
|
Distance
|
220000 |
11.4500
0.379935 minDCF
|
|
AliMeeting Overlap
AliMeeting
|
#21/24
|
Overlap
|
165000 |
11.6200
0.584293 minDCF
|
|
VoxKnesset
VoxKnesset
|
#16/24
|
Aging
|
158312 |
6.6861
0.282060 minDCF
|
|
VoxPopuli Aging
VoxPopuli
|
#17/24
|
Aging
|
146575 |
2.9644
0.143565 minDCF
|