Confidence measures for semi-automatic labeling of dialog acts

Pavel Kral 1 Christophe Cerisara 1 Jana Kleckova 2
1 PAROLE - Analysis, perception and recognition of speech
INRIA Lorraine, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : This paper deals with semi-supervised classifier training for automatic Dialog Acts (DAs) recognition. In our previous works, we have designed a dialog act recognition system for reservation applications in the Czech language. In this work, we propose to retrain this system on another corpus, for another task (broadcast news speech), in a different language (French) and with another set of dialog acts. This is realized using a semi-supervised approach based on the Expectation-Maximization (EM) algorithm. We show that, in the proposed experimental setup, the use of confidence measures to filter out incorrectly recognized dialog acts is required to improve the results. Two confidence measures are thus proposed and evaluated on the French broadcast news corpus. Experimental results confirm the interest of this approach for the task of training automatic dialog act classifiers.
Document type :
Conference papers
Liste complète des métadonnées

https://hal.inria.fr/inria-00184469
Contributor : Christophe Cerisara <>
Submitted on : Wednesday, October 31, 2007 - 9:52:24 AM
Last modification on : Friday, February 9, 2018 - 1:20:01 PM

Identifiers

  • HAL Id : inria-00184469, version 1

Collections

Citation

Pavel Kral, Christophe Cerisara, Jana Kleckova. Confidence measures for semi-automatic labeling of dialog acts. IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP 2007, Apr 2007, Honolulu, United States. ⟨inria-00184469⟩

Share

Metrics

Record views

227