Initial considerations in building a speech-to-speech translation system for the Slovenian-English language pair

EAMT 2005 Conference Proceedings 288 Initial Considerations in Building a Speech-to-Speech Translation System for the Slovenian-English Language Pair J. Žganec Gros1, A. Mihelič1, M. Žganec1, F. Mihelič2, S. Dobrišek2, J. Žibert2, Š. Vin- tar2, T. Korošec2, T. Erjavec3, M. Romih4 1Alpineon R&D, Ulica Iga Grudna 15, SI-1000 Ljubljana, Slovenia 2University of Ljubljana, SI-1000 Ljubljana, Slovenia 3Jožef Stefan Institute, Jamova 39, SI-1000 Ljubljana, Slovenia 4Amebis, Bakovnik 3, SI-1241 Kamnik, Slovenia E-mail: jerneja@alpineon.com Abstract. The paper presents the design concept of the VoiceTRAN Communicator that in- tegrates speech recognition, machine translation and text-to-speech synthesis using the DARPA Galaxy architecture. The aim of the project is to build a robust multimodal speech-to-speech translation communicator able to translate simple domain-specific sentences in the Slove- nian-English language pair. The project represents a joint collaboration between several Slovenian research organizations that are active in human language technologies. We pro- vide an overview of the task, describe the system architecture and individual servers. Fur- ther we describe the language resources that will be used and developed within the project. We conclude the paper with plans for evaluation of the VoiceTRAN Communicator. 1. Introduction Automatic speech-to-speech (STS) translation systems aim to facilitate communication among people who speak in different languages (Lavie et al., 1997), (Wahlster, 2000), (Lavie et al., 2002). Their goal is to generate a speech signal in the target language that conveys the linguis- tic information contained in the speech signal from the source language. There are, however, major open research is- sues that challenge the deployment of natural and unconstrained speech-to-speech translation systems, even for very restricted application domains, due to the fact that state-of-the-art auto- matic speech recognition and machine transla- tion systems are far from perfect. Additionally, in comparison to translating written text, con- versational spoken messages are often conveyed with imperfect syntax and casual spontaneous speech. In practice, when building demonstra- tion systems, STS systems are typically imple- mented by imposing strong constraints on the application domain and the type and structure of possible utterances, i.e. both in the range and in the scope of the user input allowed at any point of the interaction. Consequently, this com- promises the flexibility and naturalness of using the system. The VoiceTRAN Communicator is being built within a national Slovenian research project in- volving 5 partners: Alpineon, the University of Ljubljana (Faculty of Electrical Engineering, Faculty of Arts and Faculty of Social Studies), the Jožef Stefan Institute, and Amebis as a sub- contractor. The project is co-funded by the Slovenian Ministry of Defense. The aim of the project is to build a robust multimodal speech-to-speech trans- lation communicator, similar to Phraselator (Sa- rich, 2001) or Speechalator (Waibel, 2003), able to translate simple sentences in a Slovenian-Eng- lish language pair. The application domain is limited to common application scenarios that occur in peace-keeping operations on foreign missions when the users of the system have to communicate with the lo- cal population. More complex phrases can be entered via keyboard using a graphical user in- terface. Initial …

Content truncated to 500 words