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

Spoken Language Understanding research benchmarks – Page 2 · SOTA2 Research

Spoken Language Understanding

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

86.02Intent Accuracy

4

Jun 23, 2026

Spoken StoryCloze (TSTORYCLOZE)

90.2Accuracy

4

Feb 26, 2026

Spoken StoryCloze (SSTORYCLOZE)

79.9Accuracy

4

Feb 26, 2026

FluentSpeech (test)

8.3HypER

4

Feb 26, 2026

SNIPS Far Field

74.64EM Accuracy (Hyp)

4

Feb 26, 2026

SNIPS Close Field

84.88EM Accuracy (Hyp)

4

Feb 26, 2026

Internal Music dataset Full (test)

-1.16SemER

4

Feb 26, 2026

Facebook TOP recorded speech (test)

70.6EM Accuracy

3

Feb 26, 2026

internal books dataset (test)

3.31SemER

3

Feb 26, 2026

73.48Acc20

2

Feb 26, 2026

SmartLights Far field (5-fold cross-validation)

Google Dialogflow

86.25Intent Classification F1

2

Feb 26, 2026

SmartLights Close field (5-fold cross-validation)

91.72Intent Classification F1

2

Feb 26, 2026

Page 2 of 2

Popular datasets

SLURP Snips FSC ATIS SmartLights

Follow for updates

Join SOTA2 Contact