An examination of mechanisms to combine speech and sound for data analysis
Stockman, Dr. Tony
MetadataShow full item record
Speech based screen reader technology provides the most com- mon means for visually impaired users to access spreadsheets and databases. Such a spoken word interface can be effective if rela- tively small amounts of data are involved. As the size of the data set grows, it becomes increasingly difficult to carry out data anal- ysis tasks with speech alone. As part of an ongoing study into the use of data sonification to address this problem, we have intro- duced a sonification cursor, together with limited sonification of text, as mechanisms for helping to integrate the use of speech and sound. This paper and associated demonstration explore these mech- anisms further, by examining the motivation for the approach, ana- lysing its strengths and weaknesses in the context of use and ex- amining the scope for further work. The paper begins by explain- ing why speech output is the preferred means by which visually impaired users access many applications. Some recent, relevant developments in speech based screen reader systems are then de- scribed. The nature of the problems that arise in the use of spread- sheets and databases as the size of the data set increases are ex- plained. Our approach to the use of sonification in assisting access to spreadsheets is then briefly outlined. The specific mechanisms of relevance to this workshop, the sonification cursor and text soni- fication mechanisms are then explained, and examples given of their use in support of specific data analysis tasks. The paper con- cludes with a review of what has been achieved, and some direc- tions for future work.