System Detail
iic/campplus
campplus
Submitted by: Public checkpoint
- Global position
- #20 / 23
- Scenario-macro EER
- 15.1677
- Scenario-macro minDCF
- 0.590178
- 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:
iic/campplus - Submitted by: Public checkpoint
- Metadata: OpenSVBench model record
- Source: IIC
- Architecture: CAMPPlus
- Model family: CAMPPlus
- Submitted alias:
speech_campplus_sv_zh-cn_16k-common - Created at:
2026-05-29T14:29:45.445306+00:00
Training And Links
Training data and references
- Training data: Large Chinese speaker corpus, about 200k speakers
- Training setup: Official ModelScope CAMPPlus release.
- Submission mode:
full-core - Model source: Open link
- Paper: CAM++: A Fast and Efficient Network for Speaker Verification Using Context-Aware Masking
Higher-Ranked Scenarios
Scenario ranks above the global position
Show 3 supporting trial sets
- CN-Celeb: #2 on its trial-set ranking.
- CN-Celeb Genre: #3 on its trial-set ranking.
- CN-Celeb Short: #4 on its trial-set ranking.
Lower-Ranked Scenarios
Scenario ranks below the global position
Show 3 supporting trial sets
- VOiCES Noise/Reverb: #23 on its trial-set ranking.
- GLOBE: #22 on its trial-set ranking.
- VoxCeleb Short: #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 |
|---|---|---|---|---|
|
Genre-Shift Robustness
Whether performance holds when CN-Celeb enrollment and test speech come from different source genres.
|
#3/23
Top 3
|
Source-genre variation
2.6090 EER from leader
|
16.8815
0.610655 minDCF
|
|
|
Channel / Device Robustness
How resilient the model is to device and channel mismatch.
|
#13/23
Needs work
|
Device and channel variation
10.4585 EER from leader
|
14.4608
0.722188 minDCF
|
|
|
Short-Duration Robustness
How much performance holds up when speech evidence is limited by duration.
|
#13/23
Needs work
|
Short-duration speech
5.7710 EER from leader
|
14.8056
0.554588 minDCF
|
|
|
In-The-Wild Robustness
How strong the model is on unconstrained celebrity and media speech across official CN-Celeb and VoxCeleb protocols.
|
#14/23
Needs work
|
Open-domain media speech
3.0993 EER from leader
|
6.8211
0.361956 minDCF
|
|
|
Cross-Lingual Robustness
How well speaker identity survives enrollment-test language mismatch.
|
#15/23
Needs work
|
Language mismatch
4.8137 EER from leader
|
6.4713
0.346271 minDCF
|
|
|
Speaking-Style Robustness
Whether the model can preserve identity across emotion-driven change, whispered speech, and noise-induced Lombard speaking style.
|
#19/23
Needs work
|
Speaking style shift
5.6527 EER from leader
|
7.5693
0.424138 minDCF
|
|
|
Aging Robustness
How stable identity representations remain across age-derived and longitudinal recording gaps.
|
#19/23
Needs work
|
Speaker aging and time gaps
10.5459 EER from leader
|
12.3376
0.625277 minDCF
|
|
|
Distance Robustness
How much performance changes across meeting and domestic distance mismatch.
|
#19/23
Needs work
|
Distance mismatch
11.1921 EER from leader
|
16.4489
0.675161 minDCF
|
|
|
Overlap Robustness
How well speaker identity survives light, mid, and heavy overlap in both meeting and domestic recordings.
|
#20/23
Needs work
|
Overlapping speakers
10.5278 EER from leader
|
18.6371
0.617367 minDCF
|
|
|
Accent / Dialect Robustness
How reliably the model tracks identity across accent and dialect mismatch.
|
#20/23
Needs work
|
Accent and dialect variation
12.1098 EER from leader
|
28.9315
0.816441 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.
|
#23/23
Needs work
|
Noise and reverberation
21.2200 EER from leader
|
23.4800
0.737920 minDCF
|
Variant Compare
Compare CAMPPlus variants
Expand this section to compare models in the same model family by architecture, source, training data, training setup, and leaderboard result.
Show
Compare CAMPPlus 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 |
|---|---|---|---|---|---|---|---|---|
|
wespeaker/campplus-voxceleb
CAMPPlus
|
Public checkpoint | WeSpeaker | VoxCeleb2 dev | CAMPPlus with TSTP pooling, ArcMargin large-margin fine-tuning, and MUSAN/RIRS augmentation. |
#14
ranked
|
13.4165 | 0.520675 | Open |
|
iic/campplus
CAMPPlus
Current
|
Public checkpoint | IIC | Large Chinese speaker corpus, about 200k speakers | Official ModelScope CAMPPlus release. |
#20
ranked
|
15.1677 | 0.590178 | 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
|
#15/23
|
Cross-lingual
|
200000 |
6.4713
0.346271 minDCF
|
|
HI-MIA
HI-MIA
|
#9/23
|
Short-duration
|
660000 |
5.8722
0.341062 minDCF
|
|
GSC Short
Google Speech Commands
|
#5/23
|
Short-duration
|
220000 |
7.0055
0.425695 minDCF
|
|
GLOBE
GLOBE
|
#22/23
|
Accent/dialect
|
56848 |
27.7090
0.634462 minDCF
|
|
CN-Celeb
CN-Celeb
|
#2/23
|
In-the-wild
|
3484292 |
4.7029
0.224112 minDCF
|
|
CN-Celeb Genre
CN-Celeb
|
#3/23
|
Genre shift
|
440000 |
16.8815
0.610655 minDCF
|
|
CN-Celeb Short
CN-Celeb
|
#4/23
|
Short-duration
|
546964 |
19.1292
0.622269 minDCF
|
|
VoxCeleb1-O
VoxCeleb1
|
#21/23
|
In-the-wild
|
37611 |
7.7917
0.488954 minDCF
|
|
VoxCeleb1-E
VoxCeleb1
|
#21/23
|
In-the-wild
|
579818 |
7.9032
0.462346 minDCF
|
|
VoxCeleb1-H
VoxCeleb1
|
#21/23
|
In-the-wild
|
550894 |
11.1228
0.548104 minDCF
|
|
VoxCeleb Short
VoxCeleb1
|
#21/23
|
Short-duration
|
394724 |
27.2155
0.829325 minDCF
|
|
3D-Speaker Device
3D-Speaker
|
#15/23
|
Channel/device
|
180000 |
20.1800
0.916793 minDCF
|
|
3D-Speaker Distance
3D-Speaker
|
#17/23
|
Distance
|
175163 |
23.4511
0.937368 minDCF
|
|
3D-Speaker Dialect
3D-Speaker
|
#20/23
|
Accent/dialect
|
180000 |
30.1540
0.998420 minDCF
|
|
FFSVC 2022 Cross-Channel
FFSVC 2022
|
#8/23
|
Channel/device
|
72000 |
8.7417
0.527583 minDCF
|
|
FFSVC 2022 Cross-Domain
FFSVC 2022
|
#7/23
|
Distance
|
66546 |
8.7533
0.539182 minDCF
|
|
Whisper40 Whisper
Whisper40
|
#13/23
|
Speaking style
|
17600 |
11.2625
0.683063 minDCF
|
|
Lombard Grid Lombard
Lombard Grid
|
#19/23
|
Speaking style
|
29524 |
1.7884
0.119188 minDCF
|
|
VOiCES Noise/Reverb
VOiCES
|
#23/23
|
Noise/reverb
|
55000 |
23.4800
0.737920 minDCF
|
|
CHiME-6 Domestic Far-Field
CHiME-6
|
#21/23
|
Distance
|
36487 |
24.2086
0.902834 minDCF
|
|
CHiME-6 Overlap
CHiME-6
|
#20/23
|
Overlap
|
39600 |
28.3056
0.837667 minDCF
|
|
ESD
ESD
|
#20/23
|
Speaking style
|
437408 |
9.6571
0.470165 minDCF
|
|
AliMeeting Near/Far
AliMeeting
|
#16/23
|
Distance
|
220000 |
9.3825
0.321260 minDCF
|
|
AliMeeting Overlap
AliMeeting
|
#11/23
|
Overlap
|
165000 |
8.9687
0.397067 minDCF
|
|
VoxKnesset
VoxKnesset
|
#19/23
|
Aging
|
158312 |
15.4369
0.696994 minDCF
|
|
VoxPopuli Aging
VoxPopuli
|
#19/23
|
Aging
|
146575 |
9.2383
0.553561 minDCF
|