Hierarchical Finite-State Models for Speech Translation Using Categorization of Phrases
- 1.4k Downloads
In this work a hierarchical translation model is formally defined and integrated in a speech translation system. As it is well known, the relations between two languages are better arranged in terms of phrases than in terms of running words. Nevertheless phrase-based models may suffer from data sparsity at training time. The aim of this work is to improve current speech translation systems by integrating categorization within the translation model. The categories are sets of phrases either linguistically or statistically motivated. Both category and translation and acoustic models are within the framework of finite-state models. In what temporal cost is concerned, finite-state models count on efficient decoding algorithms. Regarding the spatial cost, all the models where integrated on-the-fly at decoding time, allowing an efficient use of memory.
KeywordsMachine Translation Target Language Acoustic Model Source Language Translation Model
Unable to display preview. Download preview PDF.
- 1.Callison-Burch, C., Fordyce, C., Koehn, P., Monz, C., Schroeder, J.: (meta-) evaluation of machine translation. In: Proceedings of the Second Workshop on Statistical Machine Translation, Prague, Czech Republic, pp. 136–158. Association for Computational Linguistics (2007)Google Scholar
- 2.Zhou, B., Chen, S., Gao, Y.: Constrained Phrase-based Translation Using Weighted Finite State Transducer. In: Proc. IEEE International Conference on Acoustics, Speech, and Signal Processing, vol. 1, pp. 1017–1020 (2005)Google Scholar
- 3.Pérez, A., Torres, M.I., Casacuberta, F.: Speech translation with phrase based stochastic finite-state transducers. In: Proceedings of the IEEE 32nd International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2007), Honolulu, Hawaii, USA, vol. IV, pp. 113–116. IEEE, Los Alamitos (2007)Google Scholar
- 5.Vidal, E., Thollard, F.C., de la Higuera, F.C., Carrasco, R.: Probabilistic finite-state machines - part II. IEEE Transactions on Pattern Analysis and Machine Intelligence(PAMI) 27, 1025–1039 (2005)Google Scholar
- 7.Mehryar Mohri, F.C.N.P., Riley, M.D.: AT&T FSM LibraryTM Finite-State Machine Library (2003), http://www.research.att.com/sw/tools/fsm
- 8.Martin, S., Ney, H., Zaplo, J.: Smoothing methods in maximum entropy language modeling. In: Proc. IEEE International Conference on Acoustics, Speech, and Signal Processing, Phoenix, AR, pp. 545–548 (1999)Google Scholar
- 9.Niesler, T.R., Woodland, P.C.: A variable-length category-based n-gram language model. In: IEEE ICASSP 1996, Atlanta, GA, vol. I, pp. 164–167. IEEE, Los Alamitos (1996)Google Scholar
- 11.Justo, R., Torres, M.I.: Phrases in category-based language models for Spanish and Basque ASR. In: Proceedings of the Interspeech 2007, Antwerp, Belgium, pp. 2377–2380 (2007)Google Scholar
- 12.Och, F.J.: An efficient method for determining bilingual word classes. In: Proceedings of EACL 1999, Bergen, Norway, ACL, pp. 71–76 (1999)Google Scholar