Human-system interface based on speech recognition: application to a virtual nuclear power plant control desk

This work reports the development of a human-system interface based on automatic speech recognition, and its integration to a virtual nuclear power plant control desk under development. The later is aimed to reproduce a nuclear power plant's control desk using virtual reality technology, for operator training and ergonomic evaluation purposes. Initially, users were able to interact with this virtual control desk only through computer keyboard and mouse. Then, an automatic speech recognition system was developed to serve as a new user interface, so they can navigate and operate this virtual control desk in front of a computer monitor or a projection screen, free from keyboard and mouse, through spoken commands. The automatic speech recognition interface developed is based on a well-known signal processing technique named cepstral analysis, and on artificial neural networks. The speech recognition interface is described, along with its integration with the virtual control desk, and results are presented.