Violence is a serious problems for cities like Chicago and has been exacerbated by the use of social media by gang-involved youths for taunting rival gangs. We present a corpus of tweets from a young and powerful female gang member and her communicators, which we have annotated with discourse intention, using a deep read to understand how and what triggered conversations to escalate into aggres...

We present JANUS, a speech-to-speech translation system that utilizes diverse processing strategies including connectionist learning, traditional AI knowledge representation approaches, dynamic programming, and stochastic techniques. JANUS translates continuously spoken English utterances into Japanese and German speech utterances. The overall system performance on a corpus of conference regist...

We describe the Microsoft Speech Language Translation (MSLT) corpus, which was created in order to evaluate endto-end conversational speech translation quality. The corpus was created from actual conversations over Skype, and we provide details on the recording setup and the different layers of associated text data. The corpus release includes Test and Dev sets with reference transcripts for sp...

Using instant messenger in real time communication is widespread worldwide. However, in the communication of using cross-language (Japanese other languages) instant messaging, the use of colloquial expressions usually degrades the efficiency of communication. The contributions of our research can be split into two parts: (1) we analyzed the in-house conversation logs of business correspondence ...

This paper presents a new unsupervised algorithm (WordEnds) for inferring word boundaries from transcribed adult conversations. Phone ngrams before and after observed pauses are used to bootstrap a simple discriminative model of boundary marking. This fast algorithm delivers high performance even on morphologically complex words in English and Arabic, and promising results on accurate phonetic ...

This paper describes a test collection for evaluating systems that search English SMS (Short Message Service) conversations. The collection is built from about 120,000 text messages. Topic development involved identifying typical types of information needs, then generating topics of each type for which relevant content might be found in the collection. Relevance judgments were then made for gro...

We report an empirical study on interjectory responses, utterances such as hai, ee, and un, corresponding roughly to 'yes' or '11h-huh' in English, with special emphasis on the way they help coordinate, and enhance mutual understanding in, dialogue. A corpus of data has been collected from telephone conversations, as well as from radio and television programs. Our analysis demonstrates that the...

Course books play a crucial role in educational process and are believed to be the pivotal ingredient of language teaching. Every new textbook needs to be analyzed in order to reach its best. Hence this study aims to analyze three Iranian English course books to reveal how various cultures are shown via the course books characters’ race, nationality, gender, and intercultural communications. Th...

ATR Interpreting Telecommunications Research Laboratories developed ATR-MATRIX speech translation system, which translates both ways between English and Japanese, enough to hold natural on-line real-time conversations. Using this system we started an end-to-end evaluation of a speech translation system through a dialog test with naive speakers who are not involved in system development and not ...

This paper investigates the feasibility of using crowd-sourcing services for the human assessment of machine translation quality of English and Japanese translation tasks. Nonexpert graders are hired in order to carry out a ranking-based MT evaluation of utterances taken from the domain of travel conversations. Besides a thorough analysis of the obtained non-expert grading results, data quality...