Spoken Language Understanding on Fluent Speech Commands

Metric: Accuracy (%) (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	Accuracy (%)▼	Extra Data	Paper	Date↕	Code
1	Finstreder (Conformer + AMT, character-based)	99.8	Yes	Finstreder: Simple and fast Spoken Language Unde...	2022-06-29	Code
2	UniverSLU	99.8	Yes	UniverSLU: Universal Spoken Language Understandi...	2023-10-04	-
3	Finstreder (Quartznet + AMT)	99.7	Yes	Finstreder: Simple and fast Spoken Language Unde...	2022-06-29	Code
4	textual-kd-slu	99.7	Yes	Two-stage Textual Knowledge Distillation for End...	2020-10-25	Code
5	Wav2Vec2.0-Classifier	99.7	Yes	Integration of Pre-trained Networks with Continu...	2021-04-15	-
6	E2E SLP two-step	99.7	Yes	Speech-language Pre-training for End-to-end Spok...	2021-02-11	-
7	Wav2vec 2.0 SSL	99.6	Yes	Do We Still Need Automatic Speech Recognition fo...	2021-11-29	-
8	Finstreder (Conformer)	99.5	Yes	Finstreder: Simple and fast Spoken Language Unde...	2022-06-29	Code
9	AT-AT	99.5	Yes	Exploring Transfer Learning For End-to-End Spoke...	2020-12-15	-
10	BERT, AC Pretraining	99.4	Yes	End-to-End Spoken Language Understanding for Gen...	2021-06-16	-
11	3D-CNN+LSTM+CE	99.3	Yes	Sequential End-to-End Intent and Slot Label Clas...	2021-06-08	-
12	Finstreder (Quartznet)	99.2	Yes	Finstreder: Simple and fast Spoken Language Unde...	2022-06-29	Code
13	Reptile	99.2	Yes	Improving End-to-End Speech-to-Intent Classifica...	2020-08-05	-
14	FANS	99	Yes	FANS: Fusing ASR and NLU for on-device SLU	2021-10-31	-
15	Pooling classifier pre-trained using force-aligned phoneme and word labels on LibriSpeech	98.8	Yes	Speech Model Pre-training for End-to-End Spoken ...	2019-04-07	Code
16	Amazon Alexa	98.7	Yes	Finstreder: Simple and fast Spoken Language Unde...	2022-06-29	Code
17	pGSLM+	98.2	Yes	SpeechPrompt v2: Prompt Tuning for Speech Classi...	2023-03-01	-

#1Finstreder (Conformer + AMT, character-based)SOTA
99.8
Accuracy (%)· Extra Data· 2022-06-29
Finstreder: Simple and fast Spoken Language Understanding with Finite State Transducers using modern Speech-to-Text models Code
#2UniverSLU
99.8
Accuracy (%)· Extra Data· 2023-10-04
UniverSLU: Universal Spoken Language Understanding for Diverse Tasks with Natural Language Instructions
#3Finstreder (Quartznet + AMT)
99.7
Accuracy (%)· Extra Data· 2022-06-29
Finstreder: Simple and fast Spoken Language Understanding with Finite State Transducers using modern Speech-to-Text models Code
#4textual-kd-sluSOTA
99.7
Accuracy (%)· Extra Data· 2020-10-25
Two-stage Textual Knowledge Distillation for End-to-End Spoken Language Understanding Code
#5Wav2Vec2.0-Classifier
99.7
Accuracy (%)· Extra Data· 2021-04-15
Integration of Pre-trained Networks with Continuous Token Interface for End-to-End Spoken Language Understanding
#6E2E SLP two-step
99.7
Accuracy (%)· Extra Data· 2021-02-11
Speech-language Pre-training for End-to-end Spoken Language Understanding
#7Wav2vec 2.0 SSL
99.6
Accuracy (%)· Extra Data· 2021-11-29
Do We Still Need Automatic Speech Recognition for Spoken Language Understanding?
#8Finstreder (Conformer)
99.5
Accuracy (%)· Extra Data· 2022-06-29
Finstreder: Simple and fast Spoken Language Understanding with Finite State Transducers using modern Speech-to-Text models Code
#9AT-AT
99.5
Accuracy (%)· Extra Data· 2020-12-15
Exploring Transfer Learning For End-to-End Spoken Language Understanding
#10BERT, AC Pretraining
99.4
Accuracy (%)· Extra Data· 2021-06-16
End-to-End Spoken Language Understanding for Generalized Voice Assistants
#113D-CNN+LSTM+CE
99.3
Accuracy (%)· Extra Data· 2021-06-08
Sequential End-to-End Intent and Slot Label Classification and Localization
#12Finstreder (Quartznet)
99.2
Accuracy (%)· Extra Data· 2022-06-29
Finstreder: Simple and fast Spoken Language Understanding with Finite State Transducers using modern Speech-to-Text models Code
#13ReptileSOTA
99.2
Accuracy (%)· Extra Data· 2020-08-05
Improving End-to-End Speech-to-Intent Classification with Reptile
#14FANS
99
Accuracy (%)· Extra Data· 2021-10-31
FANS: Fusing ASR and NLU for on-device SLU
#15Pooling classifier pre-trained using force-aligned phoneme and word labels on LibriSpeechSOTA
98.8
Accuracy (%)· Extra Data· 2019-04-07
Speech Model Pre-training for End-to-End Spoken Language Understanding Code
#16Amazon Alexa
98.7
Accuracy (%)· Extra Data· 2022-06-29
Finstreder: Simple and fast Spoken Language Understanding with Finite State Transducers using modern Speech-to-Text models Code
#17pGSLM+
98.2
Accuracy (%)· Extra Data· 2023-03-01
SpeechPrompt v2: Prompt Tuning for Speech Classification Tasks