Code-switched English Pronunciation Modeling for Swahili Spoken Term Detection (Pub Version, Open Access)
Abstract
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.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 03, 2016
- Accession Number
- AD1040155
Entities
People
- Charl Van Heerden
- Daniel Van Niekerk
- Marelie Davel
- Neil Kleynhans
- Rich Schwartz
- Stavros Tsakalidis
- William Hartman
Organizations
- North-West University