Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Synthetic Speech Detection research benchmarks · SOTA2 Research

Synthetic Speech Detection

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

SingFake (ID/OOD)

79.66Accuracy

27

Feb 26, 2026

92.28Accuracy

27

Feb 26, 2026

ASVspoof DF 2021 (eval)

1.88EER (%)

25

Mar 31, 2026

ASVspoof Track 1 5 (dev)

TFPARN (full model, attention pooling)

12.52EER (%)

6

Jun 4, 2026

ASVspoof LA 2019

0.35EER

6

Mar 24, 2026

In-The-Wild (ITW)

15.39EER

6

Mar 24, 2026

ASVspoof Cross-dataset Average Real-World 19/21

0.982Average AUC

5

Mar 31, 2026

0.999AUC

5

Mar 31, 2026

ASVspoof LA hidden 2021

94.4AUC

5

Mar 31, 2026

ASVspoof DF 2021 (hidden)

98.51AUC

5

Mar 31, 2026

ASVspoof 2019 (eval)

99.95AUC

5

Mar 31, 2026

Popular datasets

ASVspoof SingFake in the wild WaveFake

Follow for updates

Join SOTA2 Contact