An investigation into the use of a voice operated data input system

The aim of this study was to evaluate a voice operated data collection system intended for use in epidemiological surveys. Three groups of 45 adolescents, with similar caries experience, were examined by one examiner in schools. The caries data resulting from these examinations were collected in three ways; keyboard activated input to a portable computer, paper chart recording followed by manual input on return to base and examiner voice activated input direct to the computer. Maximum levels of accuracy were found with all three methods. The quickest method of collection in the field was by paper chart with a mean collection time per subject of 2.59 min, followed by 3.15 min for the manual keyboard method and 4.07 min for the voice activated method. When all the factors affecting time were taken into consideration it was found that overall the manual keyboard method was significantly faster than both the paper chart method and the voice activated method.