This paper describes the creation of linguistically enriched aligned corpora for Catalan, Spanish and US-English for Speech-to-Speech Translation. These corpora are obtained from two diierent sources: US-English transcribed speech data and transcriptions of conversations recorded in Catalan and Spanish. After human translation, a large trilingual spontaneous speech corpus has been obtained. Thi...

This paper gives an overview of the evaluation campaign results of the International Workshop on Spoken Language Translation (IWSLT) 20061. In this workshop, we focused on the translation of spontaneous speech. The translation directions were Arabic, Chinese, Italian, or Japanese into English. In total, 21 translation systems from 19 research groups participated in this year’s evaluation campai...

JANUS is a multi-lingual speech-tospeech translation system designed to facilitate communication between two parties engaged in a spontaneous conversation in a limited domain. In an attempt to achieve both robustness and translation accuracy we use two di erent translation components: the GLR module, designed to be more accurate, and the Phoenix module, designed to be more robust. We analyze th...

JANUS is a multilingual speech-to-speech translation system designed to facilitate communication between two parties engaged in a spontaneous conversation in a limited domain. In this paper we describe how multi-level segmentation of single utterance turns improves translation quality and facilitates accurate translation in our system. We deene the basic dialogue units that are handled by our s...

With the increasing number of applications handling spontaneous speech, the needs to process spoken languages become stronger. Speech disfluency is one of the most challenging tasks to deal with in automatic speech processing. As most applications are trained with well-formed, written texts, many issues arise when processing spontaneous speech due to its distinctive characteristics. Therefore, ...

This paper describes the method to construct and transcribe Japanese spontaneous speech data for VERBMOBIL, the German research project of speech translation.. Spontaneous spoken dialogue database is the basis for developing speech and language processing for dialogue systems such as speech translation system. The extended data of human-to-human spoken dialogue in the scenario of travel arrange...

This paper describes a new approach to language model adaptation for speech recognition based on the statistical framework of speech translation. The main idea of this approach is to compose a weighted finite-state transducer (WFST) that translates sentence styles from in-domain to out-of-domain. It enables to integrate language models of different styles of speaking or dialects and even of dif...

Some subtleties and apparent difficulties associated with the notion of spontaneous breaking of time-translation symmetry in quantum mechanics are identified and resolved. A model exhibiting that phenomenon is displayed. The possibility and significance of breaking of imaginary time-translation symmetry is discussed.

We describe the design and function of a robust processing component which is being developed for the Verbmobil speech translation system. Its task consists of collecting partial analyses of an input utterance produced by three parsers and attempting to combine them into more meaningful, larger units. It is used as a fallback mechanism in cases where no complete analysis spanning the whole inpu...

STUDY DESIGN Retrospective chart review. PURPOSE To assess whether spontaneous reduction of spondylolisthesis, as seen on magnetic resonance imaging (MRI), is related to the degree of segmental instability and low back pain. OVERVIEW OF LITERATURE The flexion-extension radiographs obtained in the sagittal plane are frequently used when segmental instability of spondylolisthesis is evaluated...