Development of multi-lingual speech recognition and text: to speech synthesis for automotive applications

In recent years speech technology has begun to be used in automotive systems for tasks such as navigation or handsfree dialling. This paper describes the development of a multi-lingual speech recognition and text-to-speech (TTS) system for the automotive environment. The system works in 9 languages: US-English, UK-English, German, French, Spanish, Italian, Dutch, Mandarin and Japanese. It is designed to work with limited resources in terms of memory size and CPU power so that it can be mounted easily on various automotive systems. The system works with diverse operating systems such as WinCE, VxWorks, Linux, pSOS and μITORN.