Section Introduction. Evaluation Inside or Assessment Outside

  • Christian Benoît
Chapter

Abstract

Some years ago, at the time when Talking Machines 1 was under its editing process, it was anticipated that standardized methods would emerge from the work carried on by international and multilingual teams of researchers. European money was then at its peak to facilitate collaboration in this area so that reference techniques would be developed, encouraged, and disseminated for wide use in the speech synthesis community. After the SPIN project, the SAM project gathered several European speech laboratories in a joint effort over eight years. SAM finally delivered a battery of tests suitable for various kinds of synthetic speech evaluation. This was also when the need for worldwide collaboration was urged, and the so-called COCOSDA2 international body was thus created to activate joint efforts among Europe, North America, Japan, and Australia. It was also at that time that a US initiative created and supported the LDC3 association. All these activities thus let Benoît and Pols [BP92, p. 436] express the wish that “in the near future, authors will systematically use multilingual assessment methods, such as those developed within the European SAM consortium, rather than publishing their work with results, more often than not, informally tested by colleagues.” How far are we from this expectation today?

Keywords

Europe Editing Acoustics Santen Diphones 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [BP92]
    C. Benoît and L. C. W. Pols. On the assessment of synthetic speech. Talking Machines: Theories, Models and Designs, G. Bailly and C. Benoît, eds. Elsevier Science Publishers B.V., North-Holland, Amsterdam, 435–441, 1992.Google Scholar
  2. [CGN92]
    R. Carlson, B. Granstrom and L. Nord. Segmental evaluation using the ESPRITSAM test procedures and mono-syllabic words. Talking Machines: Theories, Models and Designs, G. Bailly and C. Benoît, eds. Elsevier Science Publishers B.V., North-Holland, Amsterdam, 443–453, 1992.Google Scholar
  3. [Fal92]
    A. Falaschi. Segmental quality assessment by pseudo-words. Talking Machines: Theories, Models and Designs, G. Bailly and C. Benoît, eds. Elsevier Science Publishers B.V., North-Holland, Amsterdam, 455–472, 1992.Google Scholar
  4. [Fou92]
    A. Fourcin. Assessment of synthetic speech. Talking Machines: Theories, Models and Designs, G. Bailly and C. Benoît, eds. Elsevier Science Publishers B.V., North-Holland, Amsterdam, 431–434, 1992.Google Scholar
  5. [HL94]
    C. Henton and P. Litwinowicz. Saying and seeing it with feeling: Techniques for synthesizing visible, emotional speech. In Proceedings of the Second ESCA/IEEE Workshop on Speech Synthesis, New Paltz, NY, 73–76, 1994.Google Scholar
  6. [NMS94]
    S. Narayan, S. Meredith and K. Silverman. Automated speech analysis for textto-speech systems. Handout distributed at the Second ESCA/IEEE Workshop on Speech Synthesis, New Paltz, NY, 1994.Google Scholar

Copyright information

© Springer Science+Business Media New York 1997

Authors and Affiliations

  • Christian Benoît

There are no affiliations available

Personalised recommendations