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

Speech Emotion Recognition research benchmarks – Page 6 · SOTA2 Research

Speech Emotion Recognition

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

IEMOCAP non-IID (Session1–Session5 partitioning)

35.2Accuracy

4

May 26, 2026

CNN–Transformer

97.8Accuracy

4

Apr 10, 2026

CNN-Transformer

97.1Accuracy

3

Jun 11, 2026

SpectroFusion-ViT

82.19Accuracy

3

Mar 4, 2026

0WER

3

Feb 26, 2026

IEMOCAP In-Domain

73.4Weighted F1

3

Feb 26, 2026

VQ-MAE-S framework

90.2Weighted Accuracy

3

Jun 9, 2026

CNN-Transformer

98.1Accuracy

2

Jun 11, 2026

Toronto Emotional Speech Set (TESS) (test)

99Accuracy

2

Apr 30, 2026

MER 2023 (test)

89.19ACC

2

Mar 12, 2026

61.94WA

2

Feb 26, 2026

72.5WA

2

Feb 26, 2026

IEMOCAP improvised

Sun H. et al. 2022

69.1Unweighted Accuracy

2

Feb 26, 2026

eNTERFACE (full)

47.1Accuracy

2

Feb 26, 2026

65.3Accuracy

2

Feb 26, 2026

Indian regional datasets English

TEL framework (TF-EPOA + LGBM)

96.7Accuracy

1

Jun 9, 2026

Indian regional datasets Malayalam

TEL framework (TF-EPOA + LGBM)

98.42Accuracy

1

Jun 9, 2026

Indian regional datasets Tamil

TEL framework (TF-EPOA + LGBM)

99.27Accuracy

1

Jun 9, 2026

VQ-MAE-S framework

85.8Accuracy

1

Jun 9, 2026

MSP-Podcast (test)

—

—Primary metric

0

Feb 26, 2026

Page 6 of 6

Popular datasets

IEMOCAP MSP-Podcast CREMA-D RAVDESS CAFE

Follow for updates

Join SOTA2 Contact