SRI is developing a spoken language system (SLS) that should permit natural and efficient communication with an air travel information system. SLS development at SRI divides roughly into three areas: speech recognition, natural language processing, and human interface design.The paper presents an overview of SRI's development effort and an analysis of selected technical challenges in subparts of this effort, including the choice of initial domains for such technology, the architecture for the integration of the two technologies, the attributes of goal-directed spontaneous speech, and the evaluation of spoken language systems.