ROMJIST Volume 20, No. 3, 2017, pp. 210-221
Alexandru CARANICA, Horia CUCU, Andi BUZO, Corneliu BURILEANU Survey on Multilingual Spoken Term Detection
ABSTRACT: Multilinguality is a characteristic that involves the use of more than one natural language in automatic speech recognition applications. In today’s world, it is a characteristic of a rapidly increasing class of features required by a ”complete” digital assistant, through means of Automatic Speech Recognition (ASR) systems. In working environments, where more than one language is in use, the problem of storing and retrieving information acquires a multilingual dimension. Unfortunately, many languages from developing countries,orminorities,receivedverylittleattentionsofar. Onewayofimprovingthissituationis to do more research on the portability of speech and language technologies for multilingual applications, especially for under-resourced languages. These problems, as well as that of processing spoken material in a multilingual low-resource environment, will be briefly covered in this overview paper.Read full text (pdf)