Accepted Manuscript

This paper presents the evaluation of automatic break insertion for standard Basque. Basque is an agglutinative and inflected language and POS features, widely used for other languages, are not enough to accurately predict the insertion of breaks in the text. Other morpho-syntactic features, like grammatical case and information about syntagms have also been taken into account. With a textual corpus specially gathered for this study where the sentence internal punctuation marks have been removed, CARTs have been used to predict break locations. After applying parameter selection to the whole morpho-syntactic feature set, the best features were employed to build two CARTs, one that gives the same importance to deletion and insertion errors, T1, and another one, T2, that tries to minimise insertion errors. The objective evaluation of the break insertion algorithms gives a κ statistic of 0.518 and an F of 0.757 for T1 tree. The algorithms have also been subjectively evaluated and although T1 had better objective measures, the number of serious errors made by this tree is larger than the number of serious errors made by T2.

[1]  Julia Hirschberg,et al.  Automatic classification of intonational phrase boundaries , 1992 .

[2]  Stephen Cox A discriminative approach to phrase break modelling , 2005, INTERSPEECH.

[3]  Jean Carletta,et al.  Assessing Agreement on Classification Tasks: The Kappa Statistic , 1996, CL.

[4]  Ji-Hwan Kim,et al.  The use of prosody in a combined system for punctuation generation and speech recognition , 2001, INTERSPEECH.

[5]  A. Diaz De Ilarraza HIZKING21: Integrating language engineering resources and tools into systems with linguistic capa , 2003 .

[6]  Walter Daelemans,et al.  Predicting phrase breaks with memory-based learning , 2001, SSW.

[7]  Jens Apel,et al.  Have a break ! Modelling pauses in German Speech , 2004 .

[8]  Yoshihiko Gotoh,et al.  Sentence Boundary Detection in Broadcast Speech Transcripts , 2000 .

[9]  Christophe d'Alessandro,et al.  Text chunking for prosodic phrasing in French , 1998, SSW.

[10]  Eric Sanders Using Probabilistic Methods to predict Phrase Boundaries for a Text-to-Speech System , 1995 .

[11]  Walter Daelemans,et al.  Learning to Predict Pitch Accents and Prosodic Boundaries in Dutch , 2003, ACL.

[12]  Pat Langley,et al.  Selection of Relevant Features and Examples in Machine Learning , 1997, Artif. Intell..

[13]  Stephen Cox,et al.  Stochastic and syntactic techniques for predicting phrase breaks , 2007, Comput. Speech Lang..

[14]  Mari Ostendorf,et al.  A Hierarchical Stochastic Model for Automatic Prediction of Prosodic Boundary Location , 1994, CL.

[15]  Eva Navas,et al.  Assigning Phrase Breaks Using CARTs for Basque TTS , 2002 .

[16]  Yu Zheng,et al.  Using multiple linguistic features for Mandarin phrase break prediction in maximum-entropy classification framework , 2004, INTERSPEECH.

[17]  C. Clifton,et al.  Don't break, or do: prosodic boundary preferences , 2004 .

[18]  Gökhan Tür,et al.  Combining words and prosody for information extraction from speech , 1999, EUROSPEECH.

[19]  Kyuchul Yoon,et al.  A prosodic phrasing model for a Korean text-to-speech synthesis system , 2006, Comput. Speech Lang..

[20]  Helmut Schmid,et al.  New Statistical Methods for Phrase Break Prediction , 2004, COLING.

[21]  José Ignacio Hualde,et al.  A grammar of Basque , 2003 .

[22]  David B. Pisoni,et al.  Text-to-speech: the mitalk system , 1987 .

[23]  Mari Ostendorf,et al.  TOBI: a standard for labeling English prosody , 1992, ICSLP.

[24]  Ilya Oparin,et al.  Robust Rule-Based Method for Automatic Break Assignment in Russian Texts , 2005, TSD.

[25]  Paul Taylor,et al.  Assigning phrase breaks from part-of-speech sequences , 1997, Comput. Speech Lang..

[26]  Panagiotis Zervas,et al.  Bayesian induction of intonational phrase breaks , 2003, INTERSPEECH.

[27]  Julia Hirschberg,et al.  Training intonational phrasing rules automatically for English and Spanish text-to-speech , 1996, Speech Commun..

[28]  Heidi Christensen,et al.  Punctuation annotation using statistical prosody models. , 2001 .

[29]  S. Siegel,et al.  Nonparametric Statistics for the Behavioral Sciences , 2022, The SAGE Encyclopedia of Research Design.

[30]  Virach Sornlertlamvanich,et al.  Learning phrase break detection in Thai text-to-speech , 2003, INTERSPEECH.

[31]  Ren-Hua Wang,et al.  The contribution of mutual information in the intonational phrase prediction in Chinese text , 2003, International Conference on Natural Language Processing and Knowledge Engineering, 2003. Proceedings. 2003.

[32]  Jian Li,et al.  Phrase break prediction using logistic generalized linear model , 2006, INTERSPEECH.

[33]  A. Agresti An introduction to categorical data analysis , 1997 .

[34]  Sabine Buchholz,et al.  Influence of syntax on prosodic boundary prediction , 2005, INTERSPEECH.

[35]  Antonio Bonafonte,et al.  Phrase Break Prediction Using a Finite State Transducer , 2004 .

[36]  Uwe D. Reichel,et al.  Text-based and Signal-based Prediction of Break Indices and Pause Durations , 2006 .

[37]  Panagiotis Zervas,et al.  Experimental Evaluation of Tree-Based Algorithms for Intonational Breaks Representation , 2005, TSD.

[38]  M. Stone Cross‐Validatory Choice and Assessment of Statistical Predictions , 1976 .

[39]  Philipp Koehn,et al.  Improving intonational phrasing with syntactic information , 2000, 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100).

[40]  Lianhong Cai,et al.  Prosodic phrasing with inductive learning , 2002, INTERSPEECH.

[41]  Mari Ostendorf,et al.  Prosody prediction for speech synthesis using transformational rule-based learning , 1998, ICSLP.

[42]  J. M. Arriola,et al.  Combining Stochastic and Rule-Based Methods for Disambiguation in Agglutinative Languages , 1998, ACL.

[43]  Brigitte Zellner,et al.  Pauses and the temporal structure of speech , 1995 .

[44]  Panagiotis Zervas,et al.  A Data-Driven Framework for Intonational Phrase Break Prediction , 2003, TSD.

[45]  Jorge A. Gurlekian,et al.  Automatic Determination of Phrase Breaks for Argentine Spanish , 2004 .

[46]  Ren-Hua Wang,et al.  Chinese prosody phrase break prediction based on maximum entropy model , 2004, INTERSPEECH.

[47]  C. Julian Chen,et al.  Speech recognition with automatic punctuation , 1999, EUROSPEECH.

[48]  Gerard Salton,et al.  The "generality" effect and the retrieval evaluation for large collections , 1972, J. Am. Soc. Inf. Sci..

[49]  Yung-Hwan Oh,et al.  Tree-based modeling of prosodic phrasing and segmental duration for Korean TTS systems , 1999, Speech Commun..

[50]  Xuejing Sun,et al.  Intonational phrase break prediction using decision tree and n-gram model , 2001, INTERSPEECH.

[51]  P. Langley Selection of Relevant Features in Machine Learning , 1994 .

[52]  Gary Geunbae Lee,et al.  Incorporating second-order information into two-step major phrase break prediction for Korean , 2006, INTERSPEECH.

[53]  Stephen Cox,et al.  Using part-of-speech for predicting phrase breaks , 2004, INTERSPEECH.

[54]  L. Toothaker Book Review : Nonparametric Statistics for the Behavioral Sciences (Second Edition) , 1989 .

[55]  Eileen Fitzpatrick,et al.  A Computational Grammar of Discourse-Neutral Prosodic Phrasing in English , 1990, Comput. Linguistics.