Abstract
This paper proposes using a certainty-based active learning framework for extractive meeting speech summarization in order to reduce human effort in generating reference summaries. Active learning chooses a selective set of samples to be labeled by annotators. A combination of informativeness and representativeness criteria for sample selection is proposed. The results of summarizing parliamentary meeting speech show that the amount of labeled data needed for a given summarization accuracy can be reduced by more than 40 % compared to random sampling. The certainty-based active learning framework can effectively reduce the need of labeling samples for training. Furthermore, compared with lecture speech summarization task, the experiments show that the proposed active learning method of meeting speech summarization is obviously more affected by choice of different kinds of classifiers.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Fujii, Y., Yamamoto, K., Kitaoka, N., & Nakagawa, S. (2008). Class lecture summarization taking into account consecutiveness of important sentences (pp. 2438–2441). In Proceedings of Interspeech, IEEE.
Hori, C., & Furui, S. (2001). Advances in automatic speech summarization (pp. 1771–1774). In Proceedings of Eurospeech 2001.
Mrozinski, J., Whittaker, E., Chatain, P., & Furui, S. (2005). Automatic sentence segmentation of speech for automatic summarization (Vol. 1, Issue 5, p. 12). In Proceedings of ICASSP.
Kawahara, T., Nanjo, H., & Furui, S. (2001). Automatic transcription of spontaneous lecture speech (pp. 186–189). In Proceedings of the IEEE Workshop on Automatic Speech Recognition and Understanding, IEEE.
Zhang, J., Huang, S., & Fung, P. (2008). RSHMM++ for extractive lecture speech summarization (pp. 161–164). In Proceedings of 2008 I.E. Workshop on Spoken Language Technology, IEEE.
Zhang, J., & Fung, P. (2009). Active learning of extractive reference summaries for lecture speech summarization (pp. 23–26). In Proceedings of the 2nd Workshop on Building and Using Comparable Corpora (BUCC), Association for Computational Linguistics.
Schohn, G., & Cohn, D. (2000). Less is more: Active learning with support vector machines (pp. 839–846). In Machine Learning-International Workshop THEN Conference.
Tong, S., & Koller, D. (2002). Support vector machine active learning with applications to text classification. The Journal of Machine Learning Research, 2(1), 45–66.
Lewis, D., & Catlett, J. (1994). Heterogeneous uncertainty sampling for supervised learning. In Proceedings of the eleventh international conference on machine learning (pp. 148–156). Morgan Kaufmann.
Settles, B. (2009). Active learning literature survey (pp. 1648–1715). University of Wisconsin-Madison. Computer Sciences Technical Report.
Acknowledgements
Supported by the Natural Science Foundation of Guangdong Province of China (Grant No. S2012040007560), the Foundation of Guangdong Educational Committee (Grant No. 2012KJCX0099), and the National Natural Science Foundation of China (Grant No. 61300197).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Zhang, J., Yuan, H. (2014). A Certainty-Based Active Learning Framework of Meeting Speech Summarization. In: Wong, W.E., Zhu, T. (eds) Computer Engineering and Networking. Lecture Notes in Electrical Engineering, vol 277. Springer, Cham. https://doi.org/10.1007/978-3-319-01766-2_28
Download citation
DOI: https://doi.org/10.1007/978-3-319-01766-2_28
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-01765-5
Online ISBN: 978-3-319-01766-2
eBook Packages: EngineeringEngineering (R0)