User involvement in the early stages of the development of an aircraft warning system.

There is an increasing awareness of the importance and the benefits to be attained from consulting the end-users during system and product development. Although the rationale of utilizing the expertise of end-users in the system development life cycle appears to provide an apparently straightforward, even 'ideal' approach, there are many difficulties associated with eliciting the required knowledge from experts, both general and specific to every user group. Furthermore, many developers do not know how to involve users, or if they do, they do not utilize them to best effect. In the avionics sphere, the wealth of knowledge possessed by line pilots and flight engineers represents a vital information resource for the design of future flight-deck systems. As a specific example, this paper overviews some of the considerations which arose from working with these end-users in the early stages of the development of a warning and diagnostic system for civil aircraft. The end-goal of this particular phase of the work was the generation of guidelines for the design of the interface for the software engineers to use when building the prototype, and the methodological approach taken to achieve this is reported here.

Automatic speech recognition for disabled people.

It has been suggested that one of the most promising areas for the application of speech recognition is in helping handicapped people (Leggett and Williams, 1984). Within the last decade, many improvements have been made in the performance of automatic speech recognisers and current technology is discussed in relation to the needs of the disabled population. Recent research developments in the field of automatic speech recognition are reviewed, with particular reference to voice control of robotic arms and environmental control units. This includes a description of a Voice Activated Domestic Appliance System (VADAS), whose evaluation has just been completed. The general conclusion reached is that although speech recognition applications for disabled people are well within the capacity of available technology, it is primarily a lack of human factors work which is impeding developments in this field. Several human factors issues are identified; the most important of these being the need to increase the reliability of present speech recognisers, before they can confidently be incorporated into the lives of the disabled population.