Spoken Language Understanding without Speech Recognition
Authors:
- 1 April 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 6189-6193
Abstract
While conventional approaches to spoken language understanding involve cascading a speech recognizer with a language understanding system, in this paper, we describe a novel approach for deriving semantics directly from the speech signal without the need for an explicit speech recognition step. We evaluate this approach in the context of a customer care dialog system and demonstrate its effectiveness in comparison to the conventional approach.Keywords
This publication has 15 references indexed in Scilit:
- End-to-end ASR-free keyword search from speech2017
- Batch normalized recurrent neural networks2016
- End-to-end attention-based large vocabulary speech recognition2016
- Spoken Content Retrieval—Beyond Cascading Speech Recognition with Text RetrievalIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2015
- Using Recurrent Neural Networks for Slot Filling in Spoken Language UnderstandingIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2014
- Speech recognition with deep recurrent neural networks2013
- Speech-Centric Information Processing: An Optimization-Oriented ApproachProceedings of the IEEE, 2013
- Use of kernel deep convex networks and end-to-end learning for spoken language understanding2012
- An Integrative and Discriminative Technique for Spoken Utterance ClassificationIEEE Transactions on Audio, Speech, and Language Processing, 2008
- Connectionist temporal classification2006