Recent Advances in End-to-End Spoken Language Understanding
Résumé
This work deals with spoken language understanding (SLU) systems in the scenario when the semantic information is extracted directly from the audio speech signal by means of a single end-to-end neural network model. We consider two SLU tasks: named entity recognition (NER) and semantic slot filling (SF). For these tasks, in order to improve the model performance, we explore various strategies including speaker adaptive training and sequential pretraining schemes.
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|
Loading...