As speech I/O technology develops and improves, there is an increased need for standardized methods to systematically assess the performance of these systems. At the present time, speech synthesis and speech recognition technologies are at different levels of maturation and, accordingly, the procedures for testing the performance of these systems are at different stages of development. In the present paper, we describe the results of testing several text-to-speech systems using traditional intelligibility measures. In addition, we outline the design and philosophy of an automated testing procedure for measuring the performance of isolated utterance speaker-dependent speech recognition systems.
Get full access to this article
View all access options for this article.
References
1.
DoddingtonG. R.SchalkT. B. (1981). Speech recognition: Turning theory into practice. Spectrum, September, 26–32.
2.
GreeneB. G.LoganJ. S.PisoniD. B. (1986). Perception of synthetic speech produced automatically by rule: Intelligibility of eight text–to–speech systems. Behavior Research Methods, Instruments, & Computers, 18, 100–107.
3.
GreeneB. G.ManousL. M.PisoniD. B. (1984). Perceptual evaluation of DECTalk: A final report on Version 1.8. Research on Speech Perception Progress Report No. 10, Bloomington, IN: Speech Research Laboratory, Indiana University.
4.
HouseA. S.WilliamsC. E.HeckerM. H. L.KryterK. (1965). Articulation testing methods: Consonant differentiation with a closed-response set. Journal of the Acoustical Society of America, 37, 158–166.
5.
NusbaumH. C.PisoniD. B. (1985). Constraints on the perception of synthetic speech generated by rule. Behavior Research Methods, Instruments, & Computers, 17235–242.
6.
PallettD. S. (1985). Performance assessment of automatic speech recognizers. Journal of Research of the National Bureau of Standards, 90, 371–387.
