System Detail
wespeaker/res34-cnceleb
wespeaker_r34
Submitted by: Public checkpoint
- Global position
- #22 / 25
- Scenario-macro EER
- 14.4910
- Scenario-macro minDCF
- 0.607964
- Trial-set coverage
- 26 / 26
- Scenario top-three
- 1
Publicly ranked
This result is included in the reviewed public ranking.
Model Provenance
Source and architecture
- Displayed name:
wespeaker/res34-cnceleb - Submitted by: Public checkpoint
- Metadata: OpenSVBench model record
- Source: WeSpeaker
- Architecture: ResNet
- Model family: ResNet
- Submitted alias:
wespeaker-cnceleb-resnet34-LM - Created at:
2026-05-29T14:26:58.126430+00:00
Training And Links
Training data and references
- Training data: CN-Celeb train
- Training setup: ResNet34 r-vector with TSTP pooling and large-margin fine-tuning on the CN-Celeb WeSpeaker recipe.
- Score processing: None
- Submission mode:
full-core - Model source: Open link
- Paper: Deep Residual Learning for Image Recognition
Higher-Ranked Scenarios
Scenario ranks above the global position
Show 3 supporting trial sets
- CN-Celeb Genre: #3 on its trial-set ranking.
- CN-Celeb Short: #3 on its trial-set ranking.
- CN-Celeb: #8 on its trial-set ranking.
Lower-Ranked Scenarios
Scenario ranks below the global position
Show 3 supporting trial sets
- 3D-Speaker Device: #23 on its trial-set ranking.
- VoxCeleb1-H: #23 on its trial-set ranking.
- VoxPopuli Aging: #23 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 |
|---|---|---|---|---|
|
Genre-Shift Robustness
Whether performance holds when CN-Celeb enrollment and test speech come from different source genres.
|
#3/25
Top 3
|
Source-genre variation
2.4555 EER from leader
|
15.6955
0.591628 minDCF
|
|
|
Short-Duration Robustness
How much performance holds up when speech evidence is limited by duration.
|
#15/25
Needs work
|
Short-duration speech
5.6953 EER from leader
|
14.2898
0.579413 minDCF
|
|
|
In-The-Wild Robustness
How strong the model is on unconstrained celebrity and media speech across official CN-Celeb and VoxCeleb protocols.
|
#18/25
Needs work
|
Open-domain media speech
5.4574 EER from leader
|
7.8434
0.404734 minDCF
|
|
|
Speaking-Style Robustness
Whether the model can preserve identity across emotion-driven change, whispered speech, and noise-induced Lombard speaking style.
|
#19/25
Needs work
|
Speaking style shift
5.1577 EER from leader
|
7.0743
0.410279 minDCF
|
|
|
Cross-Lingual Robustness
How well speaker identity survives enrollment-test language mismatch.
|
#21/25
Needs work
|
Language mismatch
5.9194 EER from leader
|
7.5770
0.460771 minDCF
|
|
|
Aging Robustness
How stable identity representations remain across age-derived and longitudinal recording gaps.
|
#21/25
Needs work
|
Speaker aging and time gaps
10.3368 EER from leader
|
12.1285
0.635341 minDCF
|
|
|
Overlap Robustness
How well speaker identity survives light, mid, and heavy overlap in both meeting and domestic recordings.
|
#22/25
Needs work
|
Overlapping speakers
8.7206 EER from leader
|
16.4122
0.671276 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.
|
#22/25
Needs work
|
Noise and reverberation
14.3800 EER from leader
|
16.6400
0.531160 minDCF
|
|
|
Accent / Dialect Robustness
How reliably the model tracks identity across accent and dialect mismatch.
|
#22/25
Needs work
|
Accent and dialect variation
10.0688 EER from leader
|
25.8303
0.825746 minDCF
|
|
|
Distance Robustness
How much performance changes across meeting and domestic distance mismatch.
|
#23/25
Needs work
|
Distance mismatch
12.2982 EER from leader
|
16.9924
0.739294 minDCF
|
|
|
Channel / Device Robustness
How resilient the model is to device and channel mismatch.
|
#23/25
Needs work
|
Device and channel variation
14.9517 EER from leader
|
18.9178
0.837962 minDCF
|
Variant Compare
Compare ResNet variants
Compare the family variants on the overall board, a Scenario, or a Trial Set.
Show
Variant Compare
Compare ResNet variants
Compare the family variants on the overall board, a Scenario, or a Trial Set.
| Variant | Submitted by | Source | Training Data | Training Setup | Rank | EER | minDCF | Open |
|---|---|---|---|---|---|---|---|---|
|
Agora Voiceprint (ResNet293)
ResNet293 + TSTP (256-d embedding)
|
agora | custom_local_torch | VoxCeleb2 dev + VoxBlink2 + 3D-Speaker train split | AAM-Softmax / ArcMargin Speed perturbation MUSAN noise + RIR reverberation augmentation Non-VAD diff-chunk Knowledge distillation | #5 | 7.8931 | 0.297508 | Open |
|
wespeaker/res152-voxceleb
ResNet
|
Public checkpoint | WeSpeaker | VoxCeleb2 dev, 5,994 speakers | ResNet152-TSTP-emb256 r-vector, ArcMargin, 150-epoch VoxCeleb2 recipe with speed perturbation and MUSAN/RIRS augmentation. | #12 | 11.2275 | 0.442748 | Open |
|
wespeaker/res293-voxceleb
ResNet
|
Public checkpoint | WeSpeaker | VoxCeleb2 dev, 5,994 speakers | ResNet293-TSTP-emb256 r-vector, large-margin fine-tuned; official card reports 28.62M parameters and 28.10G FLOPs. | #13 | 11.3130 | 0.439607 | Open |
|
wespeaker/res34-voxceleb
ResNet
|
Public checkpoint | WeSpeaker | VoxCeleb2 dev, 5,994 speakers | ResNet34-TSTP-emb256 r-vector, large-margin fine-tuned; official card reports 6.63M parameters and 4.55G FLOPs. | #15 | 12.0516 | 0.477054 | Open |
|
wespeaker/res34-cnceleb
ResNet
Current
|
Public checkpoint | WeSpeaker | CN-Celeb train | ResNet34 r-vector with TSTP pooling and large-margin fine-tuning on the CN-Celeb WeSpeaker recipe. | #22 | 14.4910 | 0.607964 | 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
|
#21/25
|
Cross-lingual
|
200000 |
7.5770
0.460771 minDCF
|
|
HI-MIA
HI-MIA
|
#19/25
|
Short-duration
|
660000 |
7.7542
0.426640 minDCF
|
|
GSC Short
Google Speech Commands
|
#11/25
|
Short-duration
|
220000 |
8.3650
0.496165 minDCF
|
|
GLOBE
GLOBE
|
#20/25
|
Accent/dialect
|
56848 |
26.3293
0.658746 minDCF
|
|
CN-Celeb
CN-Celeb
|
#8/25
|
In-the-wild
|
3484292 |
7.4729
0.345407 minDCF
|
|
CN-Celeb Genre
CN-Celeb
|
#3/25
|
Genre shift
|
440000 |
15.6955
0.591628 minDCF
|
|
CN-Celeb Short
CN-Celeb
|
#3/25
|
Short-duration
|
546964 |
17.6675
0.600899 minDCF
|
|
VoxCeleb1-O
VoxCeleb1
|
#23/25
|
In-the-wild
|
37611 |
7.0897
0.433848 minDCF
|
|
VoxCeleb1-E
VoxCeleb1
|
#23/25
|
In-the-wild
|
579818 |
7.1685
0.432123 minDCF
|
|
VoxCeleb1-H
VoxCeleb1
|
#23/25
|
In-the-wild
|
550894 |
10.3836
0.526212 minDCF
|
|
VoxCeleb Short
VoxCeleb1
|
#22/25
|
Short-duration
|
394724 |
23.3725
0.793947 minDCF
|
|
3D-Speaker Device
3D-Speaker
|
#23/25
|
Channel/device
|
180000 |
26.8967
0.998813 minDCF
|
|
3D-Speaker Distance
3D-Speaker
|
#22/25
|
Distance
|
175163 |
25.1560
0.990655 minDCF
|
|
3D-Speaker Dialect
3D-Speaker
|
#22/25
|
Accent/dialect
|
180000 |
25.3313
0.992747 minDCF
|
|
FFSVC 2022 Cross-Channel
FFSVC 2022
|
#17/25
|
Channel/device
|
72000 |
10.9389
0.677111 minDCF
|
|
FFSVC 2022 Cross-Domain
FFSVC 2022
|
#17/25
|
Distance
|
66546 |
11.0514
0.689180 minDCF
|
|
Whisper40 Whisper
Whisper40
|
#13/25
|
Speaking style
|
17600 |
9.8125
0.615313 minDCF
|
|
Lombard Grid Lombard
Lombard Grid
|
#23/25
|
Speaking style
|
29524 |
3.3905
0.196237 minDCF
|
|
VOiCES Noise/Reverb
VOiCES
|
#22/25
|
Noise/reverb
|
55000 |
16.6400
0.531160 minDCF
|
|
CHiME-6 Domestic Far-Field
CHiME-6
|
#22/25
|
Distance
|
36487 |
22.9123
0.914742 minDCF
|
|
CHiME-6 Overlap
CHiME-6
|
#21/25
|
Overlap
|
39600 |
23.6778
0.833472 minDCF
|
|
ESD
ESD
|
#21/25
|
Speaking style
|
437408 |
8.0200
0.419287 minDCF
|
|
AliMeeting Near/Far
AliMeeting
|
#15/25
|
Distance
|
220000 |
8.8500
0.362600 minDCF
|
|
AliMeeting Overlap
AliMeeting
|
#15/25
|
Overlap
|
165000 |
9.1467
0.509080 minDCF
|
|
VoxKnesset
VoxKnesset
|
#21/25
|
Aging
|
158312 |
14.3733
0.632640 minDCF
|
|
VoxPopuli Aging
VoxPopuli
|
#23/25
|
Aging
|
146575 |
9.8837
0.638041 minDCF
|