System Detail
w2v-bert2.0 + cc + ecapa-tdnn + astp + sf2
w2v-bert2.0 + cc + ecapa-tdnn + astp + sf2
Submitted by: JinghanPeng
- Global position
- #1 / 25
- Scenario-macro EER
- 6.0847
- Scenario-macro minDCF
- 0.235010
- Trial-set coverage
- 26 / 26
- Scenario top-three
- 9
Publicly ranked
This result is included in the reviewed public ranking.
Model Provenance
Source and architecture
- Displayed name:
w2v-bert2.0 + cc + ecapa-tdnn + astp + sf2 - Submitted by: JinghanPeng
- Metadata: Contributor-provided
- Source: Contributor-provided
- Contributor-provided architecture: W2V-BERT 2.0 + MFA
- Model family: W2V-BERT 2.0
- Submitted alias:
w2v-bert2.0 + cc + ecapa-tdnn + astp + sf2 - Created at:
2026-11-12T05:33:29.817940+00:00
Training And Links
Training data and references
- Training data: VoxCeleb2 dev + VoxBlink2 + CnCeleb + TidyVoiceX train + 3D-Speaker train + KeSpeech
- Training setup: SphereFace2-C Loss Function + MUSAN/RIRS augmentation
- Score processing: None
- Submission mode:
full-core - Model source: Open link
- Paper: Cross-Lingual Speaker Verification with Self-Supervised Pre-Trained Models
Higher-Ranked Scenarios
Scenario ranks above the global position
Show 3 supporting trial sets
- AliMeeting Near/Far: #1 on its trial-set ranking.
- AliMeeting Overlap: #1 on its trial-set ranking.
- 3D-Speaker Device: #1 on its trial-set ranking.
Lower-Ranked Scenarios
Scenario ranks below the global position
Show 3 supporting trial sets
- VoxCeleb Short: #9 on its trial-set ranking.
- VOiCES Noise/Reverb: #9 on its trial-set ranking.
- HI-MIA: #6 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 |
|---|---|---|---|---|
|
In-The-Wild Robustness
How strong the model is on unconstrained celebrity and media speech across official CN-Celeb and VoxCeleb protocols.
|
#1/25
Leader
|
Open-domain media speech
0.0000 EER from leader
|
2.3860
0.102132 minDCF
|
|
|
Distance Robustness
How much performance changes across meeting and domestic distance mismatch.
|
#1/25
Leader
|
Distance mismatch
0.0000 EER from leader
|
4.6942
0.223632 minDCF
|
|
|
Overlap Robustness
How well speaker identity survives light, mid, and heavy overlap in both meeting and domestic recordings.
|
#1/25
Leader
|
Overlapping speakers
0.0000 EER from leader
|
7.6917
0.398388 minDCF
|
|
|
Short-Duration Robustness
How much performance holds up when speech evidence is limited by duration.
|
#1/25
Leader
|
Short-duration speech
0.0000 EER from leader
|
8.5945
0.315268 minDCF
|
|
|
Genre-Shift Robustness
Whether performance holds when CN-Celeb enrollment and test speech come from different source genres.
|
#1/25
Leader
|
Source-genre variation
0.0000 EER from leader
|
13.2400
0.461908 minDCF
|
|
|
Accent / Dialect Robustness
How reliably the model tracks identity across accent and dialect mismatch.
|
#1/25
Leader
|
Accent and dialect variation
0.0000 EER from leader
|
15.7615
0.443685 minDCF
|
|
|
Cross-Lingual Robustness
How well speaker identity survives enrollment-test language mismatch.
|
#2/25
Top 3
|
Language mismatch
0.1945 EER from leader
|
1.8522
0.125335 minDCF
|
|
|
Speaking-Style Robustness
Whether the model can preserve identity across emotion-driven change, whispered speech, and noise-induced Lombard speaking style.
|
#2/25
Top 3
|
Speaking style shift
0.0700 EER from leader
|
1.9866
0.119754 minDCF
|
|
|
Channel / Device Robustness
How resilient the model is to device and channel mismatch.
|
#3/25
Top 3
|
Device and channel variation
0.0414 EER from leader
|
4.0075
0.235878 minDCF
|
|
|
Aging Robustness
How stable identity representations remain across age-derived and longitudinal recording gaps.
|
#5/25
Competitive
|
Speaker aging and time gaps
0.2662 EER from leader
|
2.0580
0.052769 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.
|
#9/25
Competitive
|
Noise and reverberation
2.4000 EER from leader
|
4.6600
0.106360 minDCF
|
Variant Compare
Compare W2V-BERT 2.0 variants
Compare the family variants on the overall board, a Scenario, or a Trial Set.
Show
Variant Compare
Compare W2V-BERT 2.0 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 |
|---|---|---|---|---|---|---|---|---|
|
w2v-bert2.0 + cc + ecapa-tdnn + astp + sf2
W2V-BERT 2.0 + MFA
Current
|
JinghanPeng | Contributor-provided | VoxCeleb2 dev + VoxBlink2 + CnCeleb + TidyVoiceX train + 3D-Speaker train + KeSpeech | SphereFace2-C Loss Function + MUSAN/RIRS augmentation | #1 | 6.0847 | 0.235010 | Open |
|
W2V-BERT 2.0 + Adapter-MFA (cross-device consistency FT + weight soup)
W2V-BERT 2.0 + Adapter-MFA
|
ActionPower | WeSpeaker | Init/anchor: public wespeaker W2V-BERT 2.0 + MFA checkpoint (VoxCeleb2 dev + VoxBlink2). Fine-tuning (~40k speakers): VoxCeleb2 dev, CN-Celeb1 dev + CN-Celeb2, 3D-Speaker train, AliMeeting Train (near + far), TidyVoiceX train, Omnilingual ASR corpus, WaxalNLP, and Korean speech corpora (AI-Hub). | LoRA fine-tuning + ArcFace + cross-device consistency loss (parallel multi-device recordings) + augmentation consistency + model soup with the public wespeaker checkpoint | #2 | 7.1907 | 0.288926 | Open |
|
W2V-BERT 2.0 + MHFA (VoxCeleb2 + VoxBlink2)
W2V-BERT 2.0 + MHFA
|
pengjy | WeSpeaker | VoxCeleb2 dev + VoxBlink2: 111,301 speakers, 4.83M utterances. | Frontend: facebook/w2v-bert-2.0 (600M). Backend: MHFA with LayerNorm (64 heads, 128-d compression, all 25 hidden states), 256-d embedding. Loss: sub-center ArcFace (K=3, s=32). | #4 | 7.5635 | 0.368244 | Open |
|
Dual-LoRA W2V-BERT2-Adapter-MFA (TidyVoice)
W2V-BERT 2.0 + Adapter-MFA + Dual-LoRA
|
sgqt | WeSpeaker | VoxCeleb2 dev + VoxBlink2 + TidyVoiceX training split (3,666 speakers, approximately 262k utterances). | Dual-LoRA adaptation of a frozen W2V-BERT2-Adapter-MFA speaker-verification model pre-trained on VoxCeleb2 dev and VoxBlink2. Independent speaker and language LoRA branches are injected across all 24 transformer layers, using ranks 32 and 16, respectively. A shared language-anchored discriminator with gradient reversal removes linguistic information from the speaker representation. Training uses a three-stage curriculum over three epochs, Sub-center ArcMargin loss (K=3, scale=32, margin increased from 0.2 to 0.3), AdamW, 6-second segments, and MUSAN/RIR augmentation. At inference, only the speaker branch is retained and the speaker LoRA weights can be merged into the backbone. | #7 | 9.0606 | 0.342596 | Open |
|
wespeaker/w2vbert2
W2V-BERT 2.0 + MFA
|
Public checkpoint | WeSpeaker | VoxCeleb2 dev + VoxBlink2 | W2V-BERT 2.0 speaker-verification recipe with a 24-layer ~600M Conformer frontend and 80-dimensional normalized filterbanks. The MFA backend uses all 25 hidden states with 128-dimensional Layer Adapter compression and a 256-dimensional embedding. Stage 1 freezes the pretrained frontend and trains LoRA adapters on attention query/value projections (rank 64, alpha 128) with ArcFace (margin 0.2, scale 32), 200–300-frame crops, 0.9/1.1 speed perturbation except for VoxBlink2, and MUSAN/RIR augmentation using AdamW with weight decay 1e-4. Stage 2 jointly fine-tunes the frontend and backend for four epochs with cosine decay from 1e-5 to 5e-6. Stage 3 performs large-margin fine-tuning on VoxCeleb2 only without augmentation, using 500–600-frame crops, margin 0.5, and two epochs. AS-Norm and quality-aware score calibration are applied at evaluation. | #9 | 9.9604 | 0.360767 | 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
|
#2/25
|
Cross-lingual
|
200000 |
1.8522
0.125335 minDCF
|
|
HI-MIA
HI-MIA
|
#6/25
|
Short-duration
|
660000 |
3.6515
0.192412 minDCF
|
|
GSC Short
Google Speech Commands
|
#2/25
|
Short-duration
|
220000 |
3.6750
0.218370 minDCF
|
|
GLOBE
GLOBE
|
#4/25
|
Accent/dialect
|
56848 |
24.6130
0.505437 minDCF
|
|
CN-Celeb
CN-Celeb
|
#2/25
|
In-the-wild
|
3484292 |
4.4044
0.182453 minDCF
|
|
CN-Celeb Genre
CN-Celeb
|
#1/25
|
Genre shift
|
440000 |
13.2400
0.461908 minDCF
|
|
CN-Celeb Short
CN-Celeb
|
#1/25
|
Short-duration
|
546964 |
15.7791
0.493977 minDCF
|
|
VoxCeleb1-O
VoxCeleb1
|
#2/25
|
In-the-wild
|
37611 |
0.1542
0.009360 minDCF
|
|
VoxCeleb1-E
VoxCeleb1
|
#2/25
|
In-the-wild
|
579818 |
0.3046
0.018716 minDCF
|
|
VoxCeleb1-H
VoxCeleb1
|
#2/25
|
In-the-wild
|
550894 |
0.6443
0.037355 minDCF
|
|
VoxCeleb Short
VoxCeleb1
|
#9/25
|
Short-duration
|
394724 |
11.2724
0.356315 minDCF
|
|
3D-Speaker Device
3D-Speaker
|
#1/25
|
Channel/device
|
180000 |
4.0567
0.231867 minDCF
|
|
3D-Speaker Distance
3D-Speaker
|
#1/25
|
Distance
|
175163 |
4.6378
0.265174 minDCF
|
|
3D-Speaker Dialect
3D-Speaker
|
#2/25
|
Accent/dialect
|
180000 |
6.9100
0.381933 minDCF
|
|
FFSVC 2022 Cross-Channel
FFSVC 2022
|
#3/25
|
Channel/device
|
72000 |
3.9583
0.239889 minDCF
|
|
FFSVC 2022 Cross-Domain
FFSVC 2022
|
#3/25
|
Distance
|
66546 |
4.0192
0.243857 minDCF
|
|
Whisper40 Whisper
Whisper40
|
#2/25
|
Speaking style
|
17600 |
4.1250
0.245563 minDCF
|
|
Lombard Grid Lombard
Lombard Grid
|
#5/25
|
Speaking style
|
29524 |
0.0373
0.001826 minDCF
|
|
VOiCES Noise/Reverb
VOiCES
|
#9/25
|
Noise/reverb
|
55000 |
4.6600
0.106360 minDCF
|
|
CHiME-6 Domestic Far-Field
CHiME-6
|
#5/25
|
Distance
|
36487 |
7.1149
0.257522 minDCF
|
|
CHiME-6 Overlap
CHiME-6
|
#2/25
|
Overlap
|
39600 |
11.3500
0.572056 minDCF
|
|
ESD
ESD
|
#4/25
|
Speaking style
|
437408 |
1.7975
0.111873 minDCF
|
|
AliMeeting Near/Far
AliMeeting
|
#1/25
|
Distance
|
220000 |
3.0050
0.127975 minDCF
|
|
AliMeeting Overlap
AliMeeting
|
#1/25
|
Overlap
|
165000 |
4.0333
0.224720 minDCF
|
|
VoxKnesset
VoxKnesset
|
#6/25
|
Aging
|
158312 |
2.6000
0.080781 minDCF
|
|
VoxPopuli Aging
VoxPopuli
|
#2/25
|
Aging
|
146575 |
1.5159
0.024758 minDCF
|