Human factors affecting e-commerce services
The SMADA project worked on the development of new technologies - such as automatic speech recognition (ASR) and dialogue design. Researchers set out to effectively integrate communication methods in small terminals and to automate large proportions of the directory assistance requests handled by network operators. As part of that aim, they investigated the human factor issues involved in the use of small terminals for mobile Internet access. Assessment of these issues prior to developments for e-commerce applications could be of extreme importance. The project's human factors research aimed to find the best way to combine ASR with text and graphics in a multimodal operation. The results are useful for development of future e-commerce services. They also help to understand the issues involved in the acceptability and human factors of automated information and transaction services using speech recognition technology. The human factors experiments revealed that multimodal applications should be designed anew, since deriving these applications by adapting existing graphics or speech-only services tends to result in sub-optimal designs. Furthermore, ASR systems used in multimodal form filling applications should have specific functionalities. It has also been found that users do not spontaneously understand the way in which they should combine speech and pen input, so future multimodal services need to develop standardised interfaces for such applications. Users also prefer speech input over alternative input mechanisms if a full featured keyboard is not available. These findings are likely to be incorporated in future versions of this technology, increasing its applicability and potential widespread acceptance by users across Europe.
