Abstract
Since the quality of the language model directly affects the performance of the spoken dialog system (SDS), we should use a statistical language model (LM) trained with a large amount of data that is matched to the task domain. When porting a SDS to another language, however, it is costly to re-collect a large amount of user utterances in the target language. We thus use the language resources in a source language by utilizing statistical machine translation. The main challenge in this work is to induct automatic speech recognition results collected using a speech-input system that differs from the target SDS both in the task and the target language. To select appropriate sentences to be included in the training data for the LM, we induct a spoken language understanding module of the dialog system in the source language. Experimental construction using over three million user utterances showed that it is vital to conduct a selection from the translation results.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
- 2.
Although the speech-to-speech translation system is oriented to a travel domain, the log data contains many irrelevant inputs.
- 3.
- 4.
- 5.
- 6.
This translation system has been used in our speech-to-speech translation application called “VoiceTra,” http://mastar.jp/translation/index-en.html.
- 7.
The lexicon, the acoustic model, and language model WFSTs are composed to generate a large WFST for recognition.
References
Abe, K., Sakti, S., Isotani, R., Kawai, H., Nakamura, S.: Brazilian portuguese acoustic model training based on data borrowing from other language. In: Proceedings of Interspeech, pp. 861–864 (2010)
Bulyko, I., Ostendorf, M., Stolcke, A.: Getting more mileage fromweb text sources for conversational speech language modeling using class-dependent mixtures. In: Proceedings of Human Language Technology (HLT), vol. 2, pp. 7–9 (2003)
C Liu, L.M.: Training acoustic models with speech data from different languages. In: Proceedings of Multilingual Speech and Language Processing (2006)
Dixon, P., Finch, A., Hori, C., Kashioka, H.: Investigation on the effects of asr tuning on speech translation performance. In: Proceedings of The International Workshop on Spoken Language Translation (IWSLT), pp. 167–174 (2011)
Lefèvre, F., Mairesse, F., Young, S.: Cross-lingual spoken language understanding from unaligned data using discriminative classification models and machine translation. In: Proceedings of Interspeech, pp. 78–81 (2010)
Fan, R., Chang, K., Hsieh, C., Wang, X., Lin, C.: LIBLINEAR: a library for large linear classification. J. Machine Learn. Res. 9, 1871–1874 (2008)
Goh, C., Watanabe, T., Paul, M., Finch, A., Sumita, E.: The NICT translation system for IWSLT 2010. In: Proceedings of The International Workshop on Spoken Language Translation (IWSLT), pp. 139–146 (2010)
Hsu, B., Glass, J.: Iterative language model estimation: efficient data structure and algorithms. In: Proceedings of Interspeech, pp. 841–844 (2008)
Jabaian, B., Besacier, L., Lefèvre, F.: Combination of stochastic understanding and machine translation systems for language portability of dialogue systems. In: Proceedings of International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 5612–5615 (2011)
Kim, W., Khudanpur, S.: Lexical triggers and latent semantic analysis for cross-lingual language model adaptation. ACM Trans. Asian Lang. Inf. Process. 3(2), 94–112 (2004)
Misu, T., Kawahara, T.: A bootstrapping approach for developing language model of new spoken dialogue systems by selecting web texts. In: Proceedings of Interspeech, pp. 9–12 (2006)
Misu, T., Mizukami, E., Kashioka, H., Nakamura, S., Li, H.: A bootstrapping approach for SLU portability to a new language by inducting unannotated user queries. In: Proceedings of International Conference on Acoustics, Speech and Signal Processing (ICASSP) (2012)
Nanjo, H., Oku, Y., Yoshimi, T.: Automatic speech recognition framework for multilingual audio contents. In: Proceedings of Interspeech, pp. 1445–1448 (2007)
Sarikaya, R., Gravano, A., Gao, Y.: Rapid language model development using external resources for new spoken dialog domains. In: Proceedings of International Conference on Acoustics, Speech and Signal Processing (ICASSP), vol. 1, pp. 573–576 (2005)
Schultz, T., Waibel, A.: Language-independent and language-adaptive acoustic modeling for speech recognition. Speech Commun. 35(1–2), 31–51 (2001)
Servan, C., Camelin, N., Raymond, C., Béchet, F., De Mori, R.: On the use of machine translation for spoken language understanding portability. In: Proceedings of International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 5330–5333 (2010)
Zhu, X., Rosenfeld, R.: Improving trigram language modeling with the world wide web. In: Proceedings of International Conference on Acoustics, Speech and Signal Processing (ICASSP), vol. 1, pp. 533–536 (2001)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer Science+Business Media New York
About this paper
Cite this paper
Misu, T., Matsuda, S., Mizukami, E., Kashioka, H., Li, H. (2014). Efficient Language Model Construction for Spoken Dialog Systems by Inducting Language Resources of Different Languages. In: Mariani, J., Rosset, S., Garnier-Rizet, M., Devillers, L. (eds) Natural Interaction with Robots, Knowbots and Smartphones. Springer, New York, NY. https://doi.org/10.1007/978-1-4614-8280-2_10
Download citation
DOI: https://doi.org/10.1007/978-1-4614-8280-2_10
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4614-8279-6
Online ISBN: 978-1-4614-8280-2
eBook Packages: EngineeringEngineering (R0)