In this paper a method for generating coherent texts is described. In this method only local conditions associated with sentences determine the appropriateness of a sentence at a certain point in the text. The method does not require any form of planning and it concentrates on maximizing the amount of variation of the texts generated. 1 Introduction The topic of this paper is the generation of ...

Many studies have been reported in the domain of speech-to-speech machine translation systems for travel conversation use. Therefore, a large number of travel domain corpora have become available in recent years. From a wider viewpoint, speech-to-speech systems are required for many purposes other than travel conversation. One of these is monologues (e.g., TV news, lectures, technical presentat...

Spoken monologues feature greater sentence length and structural complexity than do spoken dialogues. To achieve high parsing performance for spoken monologues, it could prove effective to simplify the structure by dividing a sentence into suitable language units. This paper proposes a method for dependency parsing of Japanese monologues based on sentence segmentation. In this method, the depen...

We present the speech corpus SMALLWorlds (Spoken Multilingual Accounts of Logically Limited Worlds), newly established and still growing. SMALLWorlds contains monologic descriptions of scenes or worlds which are simple enough to be formally describable. The descriptions are instances of content-controlled monologue: semantically “pre-specified” but still bearing most hallmarks of spontaneous sp...

This book is Mukherjee’s PhD thesis. The study is based on a 55,000 word corpus which has been put together from parts of the London-Lund Corpus (LLC) and parts of a corpus compiled by Jürgen Esser (JEC). The author has selected monologues and dialogues from LLC; JEC consists of monologues only. This review is divided into two parts. In the first part, I will briefly summarize the seven differe...

We investigated distribution of fillers (filled pauses) in the vicinity of boundaries of different strengths in Japanese monologues, to understand whether fillers may convey information about the location and the strength of boundaries. Consistent with the results of studies on Dutch monologues, fillers tend to increase as the boundary strength grows. It has also been revealed that fillers tend...

The role of visual feedback during the production of American Sign Language was investigated by comparing the size of signing space during conversations and narrative monologues for normally sighted signers, signers with tunnel vision due to Usher syndrome, and functionally blind signers. The interlocutor for all groups was a normally sighted deaf person. Signers with tunnel vision produced a g...