Second European Conference on Speech Communication and Technology

Genova, Italy
September 24-26, 1991


Quality Evaluation of Text-to-Speech Synthesizers Using Magnitude Estimation, Categorical Estimation, Pair Comparison and Reaction Time Methods

C. Delogu, P. Paoloni, P. Pocci, C. Sementina


Recent developments in the field of speech technology leead us to assume that speech synthesis techniques could offer a quality sufficient for practical applications, such as information services or aids for handicapped people (Deliege 1989). To facilitate the diffusion of these applications it is necessary to develop reliable and valid performance measures to compare different systems for different applications. Some results on comparisons among different evaluation methodologies such as categorical estimation, magnitude estimation, paired comparison, and reaction time, are reported in the paper.

Full Paper

Bibliographic reference.  Delogu, C. / Paoloni, P. / Pocci, P. / Sementina, C. (1991): "Quality evaluation of text-to-speech synthesizers using magnitude estimation, categorical estimation, pair comparison and reaction time methods", In EUROSPEECH-1991, 353-356.