Abstract This paper describes a method of collecting mulitiligual speech data for use in the compilation of spoken command vocabularies for ICT devices and services in the EU, the EFTA countries and Turkey and Russia. The resulting vocabularies will be published as a European standard, for use by industry in the production of such applications. The context of this work is the EU i2010 framework for addressing the main challenges and developments in ICT up to 2010. Index Terms : standardisation, elicitation, validation, phone discrimination 1. Introduction Most modern ICT applications offer the option of speech as a form of command and control. Although the speech interface is not the most common form of user interface, there are a number of user groups for whom it is crucial. These are users with special needs, such as children who cannot yet read or write, visually impaired users, users with reduced ability to perceive tactile stimuli, and users with limited dexterity. As technology continues to spread to all sections of society, the proportion of users who can benefit from speech driven applications is increasing. Users have to learn a specialized vocabulary before they can effectively make use of a speech interface. Uniformity in the basic interactive elements across technologies and brands increases the transference of learning between applications. Standardisation of vocabulary for spoken access to technology is greatly facilitated by this re-application of knowledge. Such a standard has been published for 5 of the largest European languages, in terms of numbers of speakers, in 2002, i.e. English, Spanish, German, Italian and French [1]. In the interest of cultural diversity, this standardisation is to be extended to a further 23 languages, including the official EU and EFTA languages. Russian and Turkish are also included owing to the large number of speakers of these languages in the EU (see Appendix I). The work is being carried out by the European Telecommunication Standards Institute (ETSI), with experts from industry, research and development, in the context of the i2010 framework for addressing the main challenges and developments in ICT up to 2010 [2].
[1]
Chris Baber,et al.
Interactive speech technology: human factors issues in the application of speech input/output to computers
,
1993
.
[2]
Sylwia Makuch,et al.
A European Information Society for growth and employment
,
2007
.
[3]
Vladimir I. Levenshtein,et al.
Binary codes capable of correcting deletions, insertions, and reversals
,
1965
.
[4]
David A. van Leeuwen,et al.
Objective and subjective evaluation of the acoustic models of a continuous speech recognition system
,
1999,
EUROSPEECH.