Transonics: a speech to speech system for English-Persian interactions

Abstract

In this paper we describe the first phase of development of our speech-to-speech system between English and Modem Persian under the DARPA Babylon program. We give an overview of the various system components: the front end ASR, the machine translation system and the speech generation system. Challenges such as the sparseness of available spoken language data and solutions that have been employed to maximize the obtained benefits from using these limited resources are examined.Efforts in the creation of the user interface and the underlying dialog management system for mediated communication are described.

Enabling effective design of multimodal interfaces for speech-to-speech translation system: An empirical study of longitudinal user behaviors over time and user strategies for coping with errors

“…In this study, we used the Transonics (Narayanan et al, 2003) system designed for mediating and logging spoken conversations between native English and Farsi speakers. Transonics is a two-way translation system with a multimodal interface employing speech, a graphical user interface (GUI), and "push-to-talk" modalities.…”

“…In the field of speech translation, while the first commercial applications are only beginning to appear now, vibrant research efforts have been underway. These include those at BBN (Kao et al, 2008), CMU (Bach et al, 2007), IBM (Gao et al, 2006), SRI (Precoda et al, 2007), and USC (Narayanan et al, 2003;Ettelaie et al, 2006).…”

“…Much like a human translator, the S2S translation system attempts to transfer "meaning" from one language to another language, such as from English to Farsi (Persian) (Narayanan et al, 2003). The process is inherently lossy.…”

Speechlinks: Robust Cross-Lingual Tactical Communication Aids

“…The system used for the study of this chapter is a Speech-to-Speech translation device that facilitates two way spoken interactions between an English speaking doctor and a Persian (Farsi) speaking patient (51). This version of the system uses a push-to-talk modality to initiate a speaking turn which has its advantages and limitations.…”

“…To validate our approach, we 19 present and compare the performance gains achieved by the proposed approach on two Automatic Speech Recognition (ASR) systems. The first system is a medium vocabulary system for doctorpatient conversations in English (14). The second system is a large vocabulary transcription system for European parliamentary speeches (15).…”