Towards spontaneous speech recognition for on-board car navigation and information systems.

摘要: Speech recognition is seen to be of great benefit in on-board car navigation systems and assistance. The command word approach will used for applications the near future since small active vocabulary hierarchical structure much easier cope with, from developers’ side. An alternative approach, using spontaneous speech input, far more complex but provides user with an interface that very intuitive has fewer restrictions. can rely upon his or her experience inter-human communication utter queries. In this paper, we describe requirements collection a continuous data base show first results obtained under different environmental conditions car.