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

Speaker Recognition research benchmarks · SOTA2 Research

Speaker Recognition

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

VoxCeleb1 (test)

Angular Prototypical

2.21EER

126

Feb 26, 2026

VoxCeleb1 original (vox1-o)

0.24EER (mean)

13

Apr 16, 2026

VoxCeleb (test)

0.94EER

11

Mar 10, 2026

VoxCeleb1 extended (vox1-e)

0.9EER (mean)

11

Feb 26, 2026

SITW (Speakers In The Wild) core-core protocol

0.77EER

9

Feb 26, 2026

DramaSR 532K 428K utterances (test)

87.79Accuracy (All)

8

Jul 3, 2026

8.9Modified RISE-eval

8

Jun 23, 2026

VoxCeleb Hard 1

0.019EER

6

Feb 26, 2026

Speaker Recognition Dataset

1ECAPA-TDNN Score

5

Feb 26, 2026

VoxCeleb B protocol 1

1.66EER

5

Feb 26, 2026

VOICES from a Distance Challenge (Evaluation Set)

3.19EER

5

Feb 26, 2026

AudioMnist (test)

Parallel feature-free Audio processing

70.51Accuracy

2

Jun 23, 2026

AudioMnist (train)

Parallel feature-free Audio processing

71.79Accuracy

2

Jun 23, 2026

Parallel feature-free Audio processing

53.14Accuracy

2

Jun 23, 2026

Parallel feature-free Audio processing

66.67Accuracy

2

Jun 23, 2026

Popular datasets

VoxCeleb AudioMNIST TI-46 SITW (Speakers In The Wild)VOICES

Follow for updates

Join SOTA2 Contact