Code-switched English Pronunciation Modeling for Swahili Spoken Term Detection (Pub Version, Open Access)
North-West University Vanderbijlpark South Africa
Pagination or Media Count:
We investigate modeling strategies for English code-switched words as found in a Swahili spoken term detection system. Code switching, where speakers switch language in a conversation, occurs frequently in multilingual environments, and typically deteriorates STD performance. Analysis is performed in the context of the IARPA Babel program which focuses on rapid STD system development for under-resourced languages. Our results show that approaches that specifically target the modeling of code-switched words, significantly improve the detection performance of these words.