ITE Technical Report
Online ISSN : 2424-1970
Print ISSN : 1342-6893
ISSN-L : 1342-6893
Search
Advanced search
OR
Browse
Search
21.51
Displaying 1-13 of 13 articles from this issue
Show all abstracts
    • |<
    • <
    • 1
    • >
    • >|
  • [title in Japanese]
    Article type: Cover
    Pages Cover1-
    Published: September 19, 1997
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.21.51.0_Cover1
    CONFERENCE PROCEEDINGS FREE ACCESS
    Download PDF (16K)
  • [title in Japanese]
    Article type: Index
    Pages Toc1-
    Published: September 19, 1997
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.21.51.0_Toc1
    CONFERENCE PROCEEDINGS FREE ACCESS
    Download PDF (57K)
  • The Influence of Speech Signal Quality on Feature Parameters of Speaker Recognition
    Hiroshi INOUE, Takeshi KUMAGAI
    Article type: Article
    Pages 1-5
    Published: September 19, 1997
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.21.51.0_1
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    Because of a vast amount of speech information in speaker recognition, it is preferable to obtain individuality information from fewer speech data on low cost technology. Though, high quality speech data is better for precise analysis, there are problems of cost not only for hardware but time for retrieval from database and for signal processing. Generally, they say that the cepstrum have a most individuality information in individual feature of speech. In this article, we investigated the influence of speech signal quality on cepstrum coefficeients and pitch for speaker recognition. Using data limited in bandwidth and/or number of bits, change of distribution of the cepstrum coefficients are examined.
    View full abstract
    Download PDF (608K)
  • Communication System for People with Physical Disability Using Multi-modal Interface
    Megumi Nakamigawa, Yoshito Mekada, Hiroshi Hasegawa, Masao Kasuga, Kaz ...
    Article type: Article
    Pages 7-10
    Published: September 19, 1997
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.21.51.0_7
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    For people with physical disability, having access to an electronic mail system (E-mail) is a very effective method of communication that enables them to keep their privacy. However, a great amount of time and effort is necessary in order to input an E-mail text on the computer. In this paper, we propose a communication system for disabled people that use a multi-modal interface composed of a voice recognizer, a pointing device, and a text composer. Our communication system provides a convenient tool to write E-mails for these people. We measured the amount of time to generate an E-mail text and the voice recognition rate using this system. These results suggest that the system improves not only the time efficiency of text composition but also the readiness of disabled people to communicate with other people.
    View full abstract
    Download PDF (384K)
  • The Design and Development of Pointing Device for People with Severe Physical Disability for Manipulating Multimedia on Personal Computer
    Takuro HATAKEYAMA, Yoshiki KOIZUMI, Yuichi UMEDA, Masatoshi UCHIO, Mas ...
    Article type: Article
    Pages 11-16
    Published: September 19, 1997
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.21.51.0_11
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    We developed a pointing device for people with severe physical disability for manipulating multimedia on personal computer. We studied the design requirements of the input device for the disabled and developed the device that can be operated by rotating movement of the head and puff & sip. The pointing device using infrared light for communication link is composed of both a pointer and a controller. This new pointing device realizes a system which the mouse cursor are moved according to the angle of movement of the head.
    View full abstract
    Download PDF (1033K)
  • New Pointing Device Incorporating the Automatic Location Compensation Feature for People with Severe Physical Disability
    Takuro HATAKEYAMA, Yoshiki KOIZUMI, Yuichi UMEDA, Masatoshi UCHIO, Mas ...
    Article type: Article
    Pages 17-22
    Published: September 19, 1997
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.21.51.0_17
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    We developed pointing device that can be operated by rotating movement of the head and puff & sip of the people with severe physical disability like the mouse on PCs. This time, we completed a new easy-to-use method with which a user can correct the location of the pointing device without delicate operation. This paper reports the new algorithm and evaluation of this pointing device for actual use.
    View full abstract
    Download PDF (774K)
  • Bi-gram constraint segment-based speech modeling : An application to the Japanese mono-syllables recognition
    S.K. Podder, S. Tazaki, S. Tsuzuki, Y. Yamada
    Article type: Article
    Pages 23-28
    Published: September 19, 1997
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.21.51.0_23
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    A new type of stochastic modeling instead of the HMM for speech recognition, named as bi-gram constraint segment-based speech modeling (BSSM) that uses the correlation between two successive VQ codewords is presented. The main trait of this proposed approach is that the model dose not employ the concept of states as in HMM, but that of segments, corresponding to the quasi-stationary elements in speech. Since the segmentation can easily be performed apart from the function of both the training and the testing processors, the model parameters of BSSM can be estimated without any iterative task which has to be unavoidably followed in the case of HMM-based approach. Through experiments for the speaker independent Japanese mono-syllables recognition, our approach shows its outperformance over the HMM-based approaches viz. bi-gram constraint HMM and temporal correlative HMM. Moreover we have succeeded to reduce tremendously the complexities on both computation and storage compared with the HMM-based approaches.
    View full abstract
    Download PDF (678K)
  • Interactive Japanese Sign Language Interface With Attention
    Kenichirou HIRAYAMA, Takayoshi TSUJIMOTO, Yoshihito NISHIBORI, Toru YA ...
    Article type: Article
    Pages 29-34
    Published: September 19, 1997
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.21.51.0_29
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    For people who are speech impaired and use Japanese sign language, it is helpful to develop an interactive Japanese sign language interface. T. Yamaguchi, M. Yoshihara etc. have proposed and realized a method for it. It used only spatial motion features of different sign language and didn't think about the shape of hands. In this paper, we propose a method which employs skill of image processing to process and analysis simple shape of hand, and combine it with the conventional method. The proposed system switches their two features with attention. We also use Fuzzy Associative Memory Organizing Unit System (FAMOUS) to get associative inference results. Our system is person independent with high recognition rate.
    View full abstract
    Download PDF (604K)
  • [title in Japanese]
    Article type: Appendix
    Pages 35-
    Published: September 19, 1997
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.21.51.0_35
    CONFERENCE PROCEEDINGS FREE ACCESS
    Download PDF (11K)
  • Lossless Image Coding Using Orthogonal Transform
    Manabu Tatesawa, Shigeo Kato
    Article type: Article
    Pages 37-42
    Published: September 19, 1997
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.21.51.0_37
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    Orthogonal transform coding techniques are generally used for encoding natural images. Orthogonal transform with large block size is effective in reducing entropy for natural images. However it is not so effective to use fixed block size in reversible Hadamard transform. The adaptive algorithms on block size should be introduced in encoding algorithm. In this paper, we propose a reversible Handamard transform conding scheme with variable block size. In our scheme, block size is determined by the variance of transform coefficients in adjacent blocks.
    View full abstract
    Download PDF (1594K)
  • Scalable Coding for a Still Image using Dithering Method
    Shinichi Hiratsuka, Shigeo Kato
    Article type: Article
    Pages 43-48
    Published: September 19, 1997
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.21.51.0_43
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    It has been needed to develop image database services which provide images with various resolution and gray levels. However, images with a few gray levels are remarkably insufficient in image quality, so it is desirable to introduce techniques for representing efficiently these images. In this paper we propose a scalable coding method for image database services applicable to various resolution and gray level image. In this point of view, the dithering method is introduced to improve image quality in early transmission stage. The coding efficiencies of dithered images are also considered.
    View full abstract
    Download PDF (1744K)
  • Categorical color areas and constant hue loci for color stimuli of the CRT and color charts
    Hideo Otsuki, Hiroyuki Toyama, Miyoshi Ayama
    Article type: Article
    Pages 49-54
    Published: September 19, 1997
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.21.51.0_49
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    Constant hue loci based on the opponent-color theory and categorical color areas in the chromaticity diagram were determined for six observers, using 92 color charts and 56 CRT color stimuli with and without gray surround. The constant hue loci and categorical color areas seem to move conjointly among different surround conditions.
    View full abstract
    Download PDF (516K)
  • [title in Japanese]
    Article type: Appendix
    Pages App1-
    Published: September 19, 1997
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.21.51.0_App1
    CONFERENCE PROCEEDINGS FREE ACCESS
    Download PDF (73K)
    • |<
    • <
    • 1
    • >
    • >|
feedback
 Top