Speech synthesis in a computer aided learning environment

CAL is not realising its true potential in spite of the powerful facilities that can be provided with microcomputers. One way of greatly improving the appeal of CAL is to incorporate speech into the human-computer interface. This would enable a CAL workstation to emulate the teacher-pupil relationship much more closely. The logical way to provide speech is by means of text-to-speech synthesis in view of its inherent flexibility and because of the simplicity of its implementation. The speech can be provided by means of a card inside the computer. At present it is difficult to show that speech provides a quantum improvement in CAL software because current commercial text-to-speech synthesisers do not have good enough intonation. However, state-of-the-art research shows that excellent quality speech is possible provided the synthesisers can be provided with the correct parameters. Undoubtedly, current research in speech synthesis will succeed in producing almost natural sounding speech within the next 5 years (1989-94) and work on incorporating speech into CAL should be proceeding to take full advantage of the breakthrough when it arrives