Kodo Keiryogaku (The Japanese Journal of Behaviormetrics)
Online ISSN : 1880-4705
Print ISSN : 0385-5481
ISSN-L : 0385-5481
Volume 22, Issue 1
Displaying 1-7 of 7 articles from this issue
  • Kikuo MAEKAWA
    1995 Volume 22 Issue 1 Pages 1-10
    Published: March 31, 1995
    Released on J-STAGE: June 28, 2010
    JOURNAL FREE ACCESS
    A brief rough sketch of the characteristics of some selected fields of speech sciences was followed by a review of the recent models of speech sounds in phonology. In the review part of the article, priority was given to the problems of the representation of speech rather than the phonological rules. Also, special attention was paid to the autosegmental phonological representation that stresses the multidinensional temporal structure of speech. The relation between the symbolic studies and the quantitative studies was discussed at the end of the whole article.
    Download PDF (1081K)
  • Kiyoshi HONDA
    1995 Volume 22 Issue 1 Pages 11-21
    Published: March 31, 1995
    Released on J-STAGE: June 28, 2010
    JOURNAL FREE ACCESS
    Human speech production involves successive actions of biological and physical mechanisms, and they generate linguistic, neurophysiological, biomecanical, and acoustic phenomena. Aspects of the complex system have been crystalized into a variety of models in search of better understanding of speech communication processes. This review reports the existing computational and conceptual models by meany researchers, that are selected from a physiological point of view. They include peripheral components, integrated articulatory systems, motor control schemas, and models of higher brain organization.
    Download PDF (1811K)
  • Hideki KASUYA
    1995 Volume 22 Issue 1 Pages 22-29
    Published: March 31, 1995
    Released on J-STAGE: June 28, 2010
    JOURNAL FREE ACCESS
    An overview of a speech synthesis system is given, with a special emphasis on hierachical structure to generate speech sounds from an input text. A speech synthesis system generally consisits of four phases: linguistic analysis of a text, phonological/phonetic processing, synthesis parameter generation, and sound synthesis. A statistical model to represent durations of phone segments and a linear time invariant system model to generate fundamental frequency contours are shown to have been successful in improving sound quality of synthetic speech sounds.
    Download PDF (1000K)
  • Kazuhiko KAKEHI
    1995 Volume 22 Issue 1 Pages 30-39
    Published: March 31, 1995
    Released on J-STAGE: June 28, 2010
    JOURNAL FREE ACCESS
    Download PDF (1674K)
  • Yoshinori SAGISAKA
    1995 Volume 22 Issue 1 Pages 40-47
    Published: March 31, 1995
    Released on J-STAGE: June 28, 2010
    JOURNAL FREE ACCESS
    This paper describes the outline of current technology and research efforts in automatic speech recognition. The technical problems are introduced in relation to the difficulties that human beings encounter when they learn a foreign language. The structure of an automatic speech recognition system and constituent technologies are explained in comparing to human capabilities. It is found that not only engineering works but also fundamental scientific studies are still required to improve the current recognition technology.
    Download PDF (1387K)
  • —By Means of Academic Subject Score Scales—
    Naoki TAIRA
    1995 Volume 22 Issue 1 Pages 48-61
    Published: March 31, 1995
    Released on J-STAGE: June 28, 2010
    JOURNAL FREE ACCESS
    The present study tried to assess the change in academic achievements of university students after 1 year from the date of the entrance examination by using JFSA Tests and the National Center Tests which measure the level of basic academic achievement aquired in upper secondary school. Every subject's results of monitor examinations for JFSA Tests and the National Center Tests, which covered 8 school years, were compared with the achievements of the former year by means of Academic Subject Score Scales. The results showed the significant decline of achievements in almost all academic subject areas, even if regression effects were taken in account. The decline of Social Studies and Natural Sciences seemed greater than that of Japanese Language, Mathematics and English Language. One of the most important trends of change was that the preexistent differences between students of arts and science courses were enlarged. It is desired that a suitable psychological explanation will be given to those phenomena by future studies.
    Download PDF (1704K)
  • —Practical Back-Translation—
    Ryozo YOSHINO, Chikio HAYASHI, Tatsuzo SUZUKI
    1995 Volume 22 Issue 1 Pages 62-79
    Published: March 31, 1995
    Released on J-STAGE: June 28, 2010
    JOURNAL FREE ACCESS
    The main objective of this paper is to explain the process of construction and translation of the questionnaire used in our prolect of cross-national comparative social surveys by the Institute of Statistical Mathematics. In the cross-national comparative surveys, it is necessary to choose or construct a set of useful question items for the comparison, and it is also important to adequately translate those items into several languages of the survey countries. There are several problems involved in the translation, and“back translation”is a technique which gives us a cue to consider those problerns. In this paper, we show the practical procedure of construction and back translation of the questionnaire in our recent survey prolect of seven nations:Japan, USA, Britain, FRG, France, Italy, and the Netherlands.
    Download PDF (2554K)
feedback
Top