TY - JOUR
T1 - Context-sensitive and role-dependent spoken language understanding using bidirectional and attention LSTMs
AU - Hori, Chiori
AU - Hori, Takaaki
AU - Watanabe, Shinji
AU - Hershey, John R.
N1 - Publisher Copyright:
Copyright © 2016 ISCA.
PY - 2016
Y1 - 2016
N2 - To understand speaker intentions accurately in a dialog, it is important to consider the context of the surrounding sequence of dialog turns. Furthermore, each speaker may play a different role in the conversation, such as agent versus client, and thus features related to these roles may be important to the context. In previous work, we proposed context-sensitive spoken language understanding (SLU) using role-dependent long short-term memory (LSTM) recurrent neural networks (RNNs), and showed improved performance at predicting concept tags representing the intentions of agent and client in a human-human hotel reservation task. In the present study, we use bidirectional and attention-based LSTMs to train a roledependent context-sensitive model to jointly represent both the local word-level context within each utterance, and the left and right context within the dialog. The different roles of client and agent are modeled by switching between role-dependent layers. We evaluated label accuracies in the hotel reservation task using a variety of models, including logistic regression, RNNs, LSTMs, and the proposed bidirectional and attentionbased LSTMs. The bidirectional and attention-based LSTMs yield significantly better performance in this task.
AB - To understand speaker intentions accurately in a dialog, it is important to consider the context of the surrounding sequence of dialog turns. Furthermore, each speaker may play a different role in the conversation, such as agent versus client, and thus features related to these roles may be important to the context. In previous work, we proposed context-sensitive spoken language understanding (SLU) using role-dependent long short-term memory (LSTM) recurrent neural networks (RNNs), and showed improved performance at predicting concept tags representing the intentions of agent and client in a human-human hotel reservation task. In the present study, we use bidirectional and attention-based LSTMs to train a roledependent context-sensitive model to jointly represent both the local word-level context within each utterance, and the left and right context within the dialog. The different roles of client and agent are modeled by switching between role-dependent layers. We evaluated label accuracies in the hotel reservation task using a variety of models, including logistic regression, RNNs, LSTMs, and the proposed bidirectional and attentionbased LSTMs. The bidirectional and attention-based LSTMs yield significantly better performance in this task.
KW - Attention LSTMs
KW - Bidirectional LSTMs
KW - Context sensitive understanding
KW - Role-dependent model
KW - Spoken language understanding
UR - http://www.scopus.com/inward/record.url?scp=84994235660&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84994235660&partnerID=8YFLogxK
U2 - 10.21437/Interspeech.2016-1171
DO - 10.21437/Interspeech.2016-1171
M3 - Conference article
AN - SCOPUS:84994235660
VL - 08-12-September-2016
SP - 3236
EP - 3240
JO - Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
JF - Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
SN - 2308-457X
T2 - 17th Annual Conference of the International Speech Communication Association, INTERSPEECH 2016
Y2 - 8 September 2016 through 16 September 2016
ER -