Multilingual Speech Processing
- Speech UNderstanding in DIALogue (Sundial)
- JANUS (JANUS)
In multilingual speech processing, systems have to deal with more than one language. Examples include speech-to-speech translation, multilingual speech input, multilingual speech output, and cross-language spoken document retrieval and question answering. The language can either be recognised automatically by a language identifier or multiple language specific input channels can be employed. An important issue are foreign proper names in speech input and speech output (synthesis), for which special provisions must be made.