ESCA - NATO/RSG 10 Workshop on Applications of Speech Technology Lautrach, Germany September 16-17, 1993

Speech Versus Keying: A Human Factors Study

Robert I. Damper, S. D. Wood

Department of Electronics and Computer Science, University of Southampton, Southampton, UK

Speech, recognition capability is generally believed to offer great benefits in human-computer interaction. This belief has not always been supported by experimental comparisons of speech and competitor input media such as keying. One study which has purportedly shown the very significant superiority of speech over keying is that of Poock (1980). This paper argues that this finding is an artifact of a methodological Saw, specifically that the command vocabulary is chosen to suit the requirements of speech input and makes little or concession to the requirements of keying. We describe experiments modelled on those of Poock but designed to overcome this flaw and demonstrate that the claimed superiority, disappears. In fact, we find that speech input is 10.6% slower (although this difference is not statistically significant) and 360% more error-prone than, keying.

Full Paper

Bibliographic reference.  Damper, Robert I. / Wood, S. D. (1993): "Speech versus keying: a human factors study", In AST-1993, 203-206.