TY - GEN
T1 - Rhetorical-state hidden markov models for extractive speech summarization
AU - Fung, Pascale
AU - Chan, Ricky Ho Yin
AU - Zhang, Justin Jian
PY - 2008
Y1 - 2008
N2 - We propose an extractive summarization system with a novel non-generative probabilistic framework for speech summarization. One of the most underutilized features in extractive summarization is rhetorical information - semantically cohesive units that are hidden in spoken documents. We propose Rhetorical-State Hidden Markov Models (RSHMMs) to automatically decode this underlying structure in speech. We show that RSHMMs give a 71.69% ROUGE-L F-measure, a 5.69% absolute increase in lecture speech summarization performance compared to the baseline system without using RSHMM. It equally outperforms the baseline system with additional discourse features, showing that our RSHMM is a more refined improvement on the conventional discourse feature.
AB - We propose an extractive summarization system with a novel non-generative probabilistic framework for speech summarization. One of the most underutilized features in extractive summarization is rhetorical information - semantically cohesive units that are hidden in spoken documents. We propose Rhetorical-State Hidden Markov Models (RSHMMs) to automatically decode this underlying structure in speech. We show that RSHMMs give a 71.69% ROUGE-L F-measure, a 5.69% absolute increase in lecture speech summarization performance compared to the baseline system without using RSHMM. It equally outperforms the baseline system with additional discourse features, showing that our RSHMM is a more refined improvement on the conventional discourse feature.
KW - Hidden Markov models
KW - Rhetorical information
KW - Speech features
KW - Spoken document summarization
UR - https://www.webofscience.com/wos/woscc/full-record/WOS:000257456703223
UR - https://openalex.org/W2097314712
UR - https://www.scopus.com/pages/publications/51449090158
U2 - 10.1109/ICASSP.2008.4518770
DO - 10.1109/ICASSP.2008.4518770
M3 - Conference Paper published in a book
SN - 1424414849
SN - 9781424414840
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 4957
EP - 4960
BT - 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP
T2 - 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP
Y2 - 31 March 2008 through 4 April 2008
ER -