Towards multilingual interoperability in automatic speech recognition

In this communication, we address multilingual interoperability aspects in speech recognition. After giving a tentative definition of multilingual interoperability, we discuss speech recognition components and their language-specific aspects. We give a sample overview of past multilingual speech recognition research and development across different speaking styles (read, prepared and conversational). The problem of adaptation to new languages is addressed. Language-independent and crosslanguage techniques for acoustic modeling provide a means to port recognition systems to new languages without language specific acoustic data. Pronunciation lexica and text material appear to be the most crucial language-dependent resources for porting. Fast porting being a step towards multilingual interoperability the ongoing efforts of producing multilingual pronunciation lexica and collecting multilingual text corpora should be extended to the largest possible number of written languages.