Effective Incorporation of Speaker Information in Utterance Encoding in Dialog
Abstract
In dialog studies, we often encode a dialog using a hierarchical encoder where each utterance is converted into an utterance vector, and then a sequence of utterance vectors is converted into a dialog vector. Since knowing who produced which utterance is essential to understanding a dialog, conventional methods tried integrating speaker labels into utterance vectors. We found the method problematic in some cases where speaker annotations are inconsistent among different dialogs. A relative speaker modeling method is proposed to address the problem. Experimental evaluations on dialog act recognition and response generation show that the proposed method yields superior and more consistent performances.
- Publication:
-
arXiv e-prints
- Pub Date:
- July 2019
- arXiv:
- arXiv:1907.05599
- Bibcode:
- 2019arXiv190705599Z
- Keywords:
-
- Electrical Engineering and Systems Science - Audio and Speech Processing;
- Computer Science - Computation and Language;
- Computer Science - Machine Learning;
- Computer Science - Sound
- E-Print:
- 8+1 pages, 3 figures, and 5 tables. Rejected by SIGDIAL 2019