Spoken Language Understanding On Fluent

المقاييس

Accuracy (%)

النتائج

نتائج أداء النماذج المختلفة على هذا المعيار القياسي

		Paper Title
Finstreder (Conformer + AMT, character-based)	99.8	Finstreder: Simple and fast Spoken Language Understanding with Finite State Transducers using modern Speech-to-Text models
UniverSLU	99.8	UniverSLU: Universal Spoken Language Understanding for Diverse Tasks with Natural Language Instructions
Finstreder (Quartznet + AMT)	99.7	Finstreder: Simple and fast Spoken Language Understanding with Finite State Transducers using modern Speech-to-Text models
Wav2Vec2.0-Classifier	99.7	Integration of Pre-trained Networks with Continuous Token Interface for End-to-End Spoken Language Understanding
textual-kd-slu	99.7	Two-stage Textual Knowledge Distillation for End-to-End Spoken Language Understanding
E2E SLP two-step	99.7	Speech-language Pre-training for End-to-end Spoken Language Understanding
Wav2vec 2.0 SSL	99.6	Do We Still Need Automatic Speech Recognition for Spoken Language Understanding?
Finstreder (Conformer)	99.5	Finstreder: Simple and fast Spoken Language Understanding with Finite State Transducers using modern Speech-to-Text models
AT-AT	99.5	Exploring Transfer Learning For End-to-End Spoken Language Understanding
BERT, AC Pretraining	99.4	End-to-End Spoken Language Understanding for Generalized Voice Assistants
3D-CNN+LSTM+CE	99.3	Sequential End-to-End Intent and Slot Label Classification and Localization
Finstreder (Quartznet)	99.2	Finstreder: Simple and fast Spoken Language Understanding with Finite State Transducers using modern Speech-to-Text models
Reptile	99.2	Improving End-to-End Speech-to-Intent Classification with Reptile
FANS	99.0	FANS: Fusing ASR and NLU for on-device SLU
Pooling classifier pre-trained using force-aligned phoneme and word labels on LibriSpeech	98.8	Speech Model Pre-training for End-to-End Spoken Language Understanding
Amazon Alexa	98.7	Finstreder: Simple and fast Spoken Language Understanding with Finite State Transducers using modern Speech-to-Text models
pGSLM+	98.2	SpeechPrompt v2: Prompt Tuning for Speech Classification Tasks

0 of 17 row(s) selected.