Topic Detection in Read Documents
This paper addresses the problem of topic annotation in the speech retrieval domain. It describes an algorithm developed to perform automatic topic annotation of broadcast news (BN) speech corpora. The adopted approach is based in Hidden Markov Models (HMM) and topic language models, solving the topic segmentation and labelling tasks simultaneously. To overcome the lack of topic labelled material for training statistical models, a two-stage unsupervised clustering was developed. Both stages are based on the nearestneighbour search method, using the Kullback-Leibler distance. On-going experiments to evaluate the system performance are also described.
Unable to display preview. Download preview PDF.
- 1.Fiscus, J., Doddington, G., Garofolo, J., Martin, A., “NIST’S 1998 Topic Detection and Tracking Evaluation (TDT2)”, in Proc. DARPA Broadcast News Workshop, Feb. 1999.Google Scholar
- 2.Ng, K., “Survey of Approaches to Information Retrieval of Speech Messages” Technical report, Spoken Language Systems Group, MIT, February 1996.Google Scholar
- 3.Yamron, J. P., Carp, I., Gillick, L., Lowe, S., “A Hidden Markov Model Approach to Text Segmentation and Event Tracking”, in Proceedings of ICASSP-98, Seattle, May 1998.Google Scholar
- 4.Clarkson, P., Rosenfeld, R., “Statistical Language Modeling using the CMU-Cambridge Toolkit”, in Proc. EUROSPEECH 97, Rhodes, Greece, 1997.Google Scholar
- 5.Neto, J., Martins, C., Meinedo, H., and Almeida, L., “The Design of a Large Vocabulary Speech Corpus for Portuguese”, in Proc. EUROSPEECH 97, Rhodes, Greece, 1997.Google Scholar
- 6.Caseiro, D., Trancoso, I., “A Decoder for Finite-State Structured Search Spaces” submitted to ISCA ITRW International Workshop on “Automatic Speech Recognition: Challenges for the Next Millennium”, Paris, France, September 18-20, 2000.Google Scholar