We describe the architecture of the Spoken Language Translator (SLT), a prototype speech translation system which can translate queries from spoken English to spoken Swedish in the domain of air travel information systems. Though the performance given the level of effort so far has been extremely encouraging, more work is needed to provide a technology that will support widespread applications....

listening skill and speaking skill, according language didactics, have close interaction in lingual communications. therefore, measuring listening comprehension without considering other language skills seems to be difficult. the present study is an effort to compare the effectiveness of oral and written multiple-choice listening tests for persian language learners. in fact, this study answers ...

This paper presents a method of document expansion using a side collection for improving the overall performance in retrieving spoken documents using text queries. This method is applied to Chinese spoken document retrieval (SDR) tasks where a series of experiments have been carried out for both monolingual and cross-language SDR systems. In our monolingual retrieval experiments, Cantonese broa...

This study investigated the impact of interaction and output modality on vocabulary learning and retention of EFL learners. To investigate the impact of Interaction, solitary (n =69) and collaborative (n =62) groups served as experimental and No Interaction No Output (n =26) as control group. To address the effect of modality, spoken (n =39) and written (n =31) modalities served as experimental...

This study investigated the impact of interaction and output modality on vocabulary learning and retention of EFL learners. To investigate the impact of Interaction, solitary (n =69) and collaborative (n =62) groups served as experimental and No Interaction No Output (n =26) as control group. To address the effect of modality, spoken (n =39) and written (n =31) modalities served as experimental...

In this paper, we exploit non-local features as an estimate of long-distance dependencies to improve performance on the statistical spoken language understanding (SLU) problem. The statistical natural language parsers trained on text perform unreliably to encode non-local information on spoken language. An alternative method we propose is to use trigger pairs that are automatically extracted by...

Automatic Detection and Classification of Prosodic Events Andrew Rosenberg Prosody, or intonation, is a critically important component of spoken communication. The automatic extraction of prosodic information is necessary for machines to process speech with human levels of proficiency. In this thesis we describe work on the automatic detection and classification of prosodic events – specificall...

As a consequence of the established practice to prefer training data obtained from written sources, NLP tools encounter problems in handling data from the spoken domain. However, accurate models of spoken data are increasingly in demand for naturalistic speech generation and machine translations in speech-like contexts (such as chat windows and SMS). There is a widely held assumption in the lin...

We present a performance evaluation framework for Spoken Language Understanding (SLU) modules, focusing on three elements: (1) characterization of spoken utterances, (2) experimental design, and (3) quantitative evaluation metrics. We then describe the application of our framework to Scusi?— our SLU system that focuses on referring expressions.