An Active Learning Approach for Statistical Spoken Language Understanding
In general, large amount of segmented and labeled data is needed to estimate statistical language understanding systems. In recent years, different approaches have been proposed to reduce the segmentation and labeling effort by means of unsupervised o semi-supervised learning techniques. We propose an active learning approach to the estimation of statistical language understanding models that involves the transcription, labeling and segmentation of a small amount of data, along with the use of raw data. We use this approach to learn the understanding component of a Spoken Dialog System. Some experiments that show the appropriateness of our approach are also presented.
Keywordsactive learning unaligned corpus spoken language understanding spoken dialog systems
- 3.Gotab, P., Bechet, F., Damnati, G.: Active learning for rule-based and corpus-based spoken labguage understanding moldes. In: IEEE Workshop Automatic Speech Recognition and Understanding (ASRU 2009), pp. 444–449 (2009)Google Scholar
- 4.Gotab, P., Damnati, G., Becher, F., Delphin-Poulat, L.: Online slu model adaptation with a partial oracle. In: Proc. of InterSpeech 2010, Makuhari, Chiba, Japan, pp. 2862–2865 (2010)Google Scholar
- 6.Ortega, L., Galiano, I., Hurtado, L.F., Sanchis, E., Segarra, E.: A statistical segment-based approach for spoken language understanding. In: Proc. of InterSpeech 2010, Makuhari, Chiba, Japan, pp. 1836–1839 (2010)Google Scholar