A Probabilistic Generative Framework for Extractive Broadcast News Speech Summarization

In this paper, we consider extractive summarization of broadcast news speech and propose a unified probabilistic generative framework that combines the sentence generative probability and the sentence prior probability for sentence ranking. Each sentence of a spoken document to be summarized is treated as a probabilistic generative model for predicting the document. Two matching strategies, namely literal term matching and concept matching, are thoroughly investigated. We explore the use of the language model (LM) and the relevance model (RM) for literal term matching, while the sentence topical mixture model (STMM) and the word topical mixture model (WTMM) are used for concept matching. In addition, the lexical and prosodic features, as well as the relevance information of spoken sentences, are properly incorporated for the estimation of the sentence prior probability. An elegant feature of our proposed framework is that both the sentence generative probability and the sentence prior probability can be estimated in an unsupervised manner, without the need for handcrafted document-summary pairs. The experiments were performed on Chinese broadcast news collected in Taiwan, and very encouraging results were obtained.

[1]  Elizabeth D. Liddy,et al.  Advances in Automatic Text Summarization , 2001, Information Retrieval.

[2]  Bin Chen,et al.  A comparative study of probabilistic ranking models for spoken document summarization , 2008, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing.

[3]  Timothy J. Hazen,et al.  Retrieval and browsing of spoken content , 2008, IEEE Signal Processing Magazine.

[4]  Hsin-Min Wang,et al.  Statistical Chinese spoken document retrieval using latent topical information , 2004, INTERSPEECH.

[5]  Lin-Shan Lee,et al.  A discriminative HMM/N-gram-based retrieval approach for mandarin spoken documents , 2004, TALIP.

[6]  Massih-Reza Amini,et al.  Automatic Text Summarization Based on Word-Clusters and Ranking Algorithms , 2005, ECIR.

[7]  Vibhu O. Mittal,et al.  Ultra-Summarization: A Statistical Approach to Generating Highly Condensed Non-Extractive Summaries (poster abstract). , 1998, SIGIR 1999.

[8]  Hsin-Min Wang,et al.  A unified probabilistic generative framework for extractive spoken document summarization , 2007, INTERSPEECH.

[9]  S. Renals,et al.  Content-based access to spoken audio , 2005, IEEE Signal Processing Magazine.

[10]  Lin-Shan Lee,et al.  IMPROVED SUMMARIZATION OF CHINESE SPOKEN DOCUMENTS BY PROBABILISTIC LATENT SEMANTIC ANALYSIS (PLSA) WITH FURTHER ANALYSIS AND INTEGRATED SCORING , 2006, 2006 IEEE Spoken Language Technology Workshop.

[11]  Konstantinos Koumpis,et al.  Automatic summarization of voicemail messages using lexical and prosodic features , 2005, TSLP.

[12]  Jean Carletta,et al.  Extractive summarization of meeting recordings , 2005, INTERSPEECH.

[13]  Ronald Rosenfeld,et al.  Whole-sentence exponential language models: a vehicle for linguistic-statistical integration , 2001, Comput. Speech Lang..

[14]  W. Bruce Croft,et al.  Query expansion using local and global document analysis , 1996, SIGIR '96.

[15]  Vibhu O. Mittal,et al.  Ultra-summarization (poster abstract): a statistical approach to generating highly condensed non-extractive summaries , 1999, SIGIR '99.

[16]  Hsin-Min Wang,et al.  Extractive Chinese Spoken Document Summarization Using Probabilistic Ranking Models , 2006, ISCSLP.

[17]  Lisa F. Rau,et al.  Automatic Condensation of Electronic Publications by Sentence Selection , 1995, Inf. Process. Manag..

[18]  Pascale Fung,et al.  Improving lecture speech summarization using rhetorical information , 2007, 2007 IEEE Workshop on Automatic Speech Recognition & Understanding (ASRU).

[19]  Berlin Chen,et al.  Word Topical Mixture Models for Extractive Spoken Document Summarization , 2007, 2007 IEEE International Conference on Multimedia and Expo.

[20]  Sadaoki Furui,et al.  Sentence extraction-based presentation summarization techniques and evaluation metrics , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[21]  Gerald Penn,et al.  Evaluation of Sentence Selection for Speech Summarization , 2005 .

[22]  Berlin Chen,et al.  Lightly supervised and data-driven approaches to Mandarin broadcast news transcription , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[23]  Pascale Fung,et al.  Speech Summarization Without Lexical Features for Mandarin Broadcast News , 2007, NAACL.

[24]  Julia Hirschberg,et al.  Summarizing Speech Without Text Using Hidden Markov Models , 2006, NAACL.

[25]  Phyllis B. Baxendale,et al.  Machine-Made Index for Technical Literature - An Experiment , 1958, IBM J. Res. Dev..

[26]  Francine Chen,et al.  A trainable document summarizer , 1995, SIGIR '95.

[27]  Julia Hirschberg,et al.  Comparing lexical, acoustic/prosodic, structural and discourse features for speech summarization , 2005, INTERSPEECH.

[28]  Berlin Chen,et al.  Speech retrieval of Mandarin broadcast news via mobile devices , 2005, INTERSPEECH.

[29]  Roberto Togneri,et al.  Prosodic features for a maximum entropy language model , 2006, INTERSPEECH.

[30]  Xin Liu,et al.  Generic text summarization using relevance measure and latent semantic analysis , 2001, SIGIR '01.

[31]  Michel Galley,et al.  A Skip-Chain Conditional Random Field for Ranking Meeting Utterances by Importance , 2006, EMNLP.

[32]  Danushka Bollegala,et al.  A Bottom-Up Approach to Sentence Ordering for Multi-Document Summarization , 2006, ACL.

[33]  Qin Lu,et al.  Extractive Summarization using Inter- and Intra- Event Relevance , 2006, ACL.

[34]  Hsin-Min Wang,et al.  Spoken document summarization using relevant information , 2007, 2007 IEEE Workshop on Automatic Speech Recognition & Understanding (ASRU).

[35]  Tatsuya Kawahara,et al.  Automatic indexing of lecture presentations using unsupervised learning of presumed discourse markers , 2004, IEEE Transactions on Speech and Audio Processing.

[36]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[37]  Lin-Shan Lee,et al.  Discriminating capabilities of syllable-based features and approaches of utilizing them for voice retrieval of speech information in Mandarin Chinese , 2002, IEEE Trans. Speech Audio Process..

[38]  Berlin Chen,et al.  Exploring the use of latent topical information for statistical Chinese spoken document retrieval , 2006, Pattern Recognit. Lett..

[39]  Frederick Jelinek,et al.  Statistical methods for speech recognition , 1997 .

[40]  Sadaoki Furui,et al.  Sentence-extractive automatic speech summarization and evaluation techniques , 2006, Speech Commun..

[41]  Thomas Hofmann,et al.  Unsupervised Learning by Probabilistic Latent Semantic Analysis , 2004, Machine Learning.

[42]  Jen-Tzung Chien,et al.  Adaptive Bayesian Latent Semantic Analysis , 2008, IEEE Transactions on Audio, Speech, and Language Processing.

[43]  Chung-Hsien Wu,et al.  Spoken document summarization using acoustic, prosodic and semantic information , 2005, 2005 IEEE International Conference on Multimedia and Expo.

[44]  Lin-shan Lee,et al.  Spoken document understanding and organization , 2005, IEEE Signal Processing Magazine.

[45]  Berlin Chen,et al.  Training data selection for improving discriminative training of acoustic models , 2007, 2007 IEEE Workshop on Automatic Speech Recognition & Understanding (ASRU).

[46]  Sadaoki Furui,et al.  Speech-to-text and speech-to-speech summarization of spontaneous speech , 2004, IEEE Transactions on Speech and Audio Processing.

[47]  Pushpak Bhattacharyya,et al.  Generic Text Summarization Using WordNet , 2004, LREC.

[48]  Chris D. Paice,et al.  Constructing literature abstracts by computer: Techniques and prospects , 1990, Inf. Process. Manag..

[49]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[50]  Berlin Chen,et al.  Chinese Spoken Document Summarization Using Probabilistic Latent Topical Information , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.

[51]  Berlin Chen,et al.  Word Topical Mixture Models for Dynamic Language Model Adaptation , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.