Abstract
Automatic story segmentation is an important prerequisite for semantic-level applications. The normalized cuts (NCuts) method has recently shown great promise for segmenting English spoken lectures. However, the availability assumption of the exact story number per file significantly limits its capability to handle a large number of transcripts. Besides, how to apply such method to Chinese language in the presence of speech recognition errors is unclear yet. Addressesing these two problems, we propose a self-validated NCuts (SNCuts) algorithm for segmenting Chinese broadcast news via inaccurate lexical cues, generated by the Chinese large vocabulary continuous speech recognizer (LVCSR). Due to the specialty of Chinese language, we present a subword-level graph embedding for the erroneous LVCSR transcripts. We regularize the NCuts criterion by a general exponential prior of story numbers, respecting the principle of Occam’s razor. Given the maximum story number as a general parameter, we can automatically obtain reasonable segmentations for a large number of news transcripts, with the story numbers automatically determined for each file, and with comparable complexity to alternative non-self-validated methods. Extensive experiments on benchmark corpus show that: (i) the proposed SNCuts algorithm can efficiently produce comparable or even better segmentation quality, as compared to other state-of-the-art methods with true story number as an input parameter; and (ii) the subword-level embedding always helps to recovering lexical cohesion in Chinese erroneous transcripts, thus improving both segmentation accuracy and robustness to LVCSR errors.
L. Xie—This work is supported by NSFC 61671325, 61572354.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
CCTV Corpus: Story segmentation and topic detection of CCTV Mandarin broadcast news (2010)
Chan, S.K., Xie, L., Meng, H.M.L.: Modeling the statistical behavior of lexical chains to capture word cohesiveness for automatic story segmentation. In: INTERSPEECH, pp. 2408–2411 (2007)
Choi, F.: Advances in domain independent linear text segmentation. In: NAACL, pp. 26–33 (2000)
Choi, F., Wiemer-Hastings, P., Moore, J.: Latent semantic analysis for story segmentation. In: EMNLP (2001)
Feng, W., Huang, W., Ren, J.: Class imbalance ensemble learning based on the margin theory. Appl. Sci. 8(5), 815 (2018)
Feng, W., Jia, J., Liu, Z.Q.: Self-validated labeling of Markov random fields for image segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 32(10), 1871–1887 (2010)
Guo, Q., Sun, S., Ren, X., Dong, F., Gao, B.Z., Feng, W.: Freqeuncy-tuned active contour model. Neurocomputing 275(31), 2307–2316 (2018)
Hearst, M.: TextTiling: segmentation text into multi-paragraph subtopic passages. Comput. Linguist. 23(1), 33–64 (1997)
Kyoto University: Multipurpose large vocabulary continuous speech recognition engine - Julius (rev 3.2) (2001)
Lee, L.S., Chen, B.: Spoken document understanding and organization. IEEE Signal Process. Mag. 22(5), 42–60 (2005)
Liu, Z., Xie, L., Feng, W.: Maximum lexical cohesion for fine-grained news story segmentation. In: INTERSPEECH (2010)
Malioutov, I., Barzilay, R.: Minimum cut model for spoken lecture segmentation. In: ACL, pp. 25–32 (2006)
Nie, X., Feng, W., Wan, L., Xie, L.: Measuring similarity by contextual word connections in Chinese news story segmentation. In: ICASSP (2013)
NIST: The topic detection and tracking phase 2 (TDT2) evaluation plan, version 35 (1998)
Ren, J., Jiang, J.: Hierarchical modeling and adaptive clustering for real-time summarization of rush videos. IEEE Trans. Multimed. 11(5), 906–917 (2009)
Shi, J., Malik, J.: Normalized cuts and image segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 22(8), 888–905 (2000)
Stokes, N., Carthy, J., Smeaton, A.: SeLeCT: a lexical cohesion based news story segmentation system. J. AI Commun. 17(1), 3–12 (2004)
Storn, R., Price, K.: Differential evolution - a simple and efficient heuristic for global optimization over continuous spaces. J. Glob. Optim. 11, 341–359 (1997)
TDT2 Corpus: Topic detection and tracking phase 2, July 2000. http://projects.ldc.upenn.edu/TDT2/
Wang, X., Xie, L., Ma, B., Chng, E.S., Li, H.: Modeling broadcast news prosody using conditional random fields for story segmentation. In: APSIPA ASC (2010)
Xie, L., Zheng, L., Liu, Z., Zhang, Y.: Laplacian Eigenmaps for automatic story segmentation of broadcast news. IEEE Trans Audio Speech Lang. Process. 20(1), 264–277 (2012)
Yang, Y., Xie, L.: Subword latent semantic analysis for texttiling-based automatic story segmentation of Chinese broadcast news. In: ISCSLP, pp. 358–361 (2008)
Zhang, J., Xie, L., Feng, W., Zhang, Y.: A subword normalized cut approach to automatic story segmentation of Chinese broadcast news. In: Lee, G.G., Song, D., Lin, C.-Y., Aizawa, A., Kuriyama, K., Yoshioka, M., Sakai, T. (eds.) AIRS 2009. LNCS, vol. 5839, pp. 136–148. Springer, Heidelberg (2009). https://doi.org/10.1007/978-3-642-04769-5_12
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Feng, W., Xie, L., Zhang, J., Zhang, Y., Zhang, Y. (2018). Self-validated Story Segmentation of Chinese Broadcast News. In: Ren, J., et al. Advances in Brain Inspired Cognitive Systems. BICS 2018. Lecture Notes in Computer Science(), vol 10989. Springer, Cham. https://doi.org/10.1007/978-3-030-00563-4_55
Download citation
DOI: https://doi.org/10.1007/978-3-030-00563-4_55
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-00562-7
Online ISBN: 978-3-030-00563-4
eBook Packages: Computer ScienceComputer Science (R0)