Van Niekerk, DRBarnard, E2008-01-242008-01-242007-11Van Niekerk, DR and Barnard, E. 2007. Important factors in HMM-based phonetic segmentation. 18th Annual Symposium of the Pattern Recognition Association of South Africa (PRASA), Pietermaritzburg, Kwazulu-Natal, South Africa, 28-30 November 2007, pp 6978-1-86840-656-2http://hdl.handle.net/10204/19782007: PRASAWhen doing research into or building systems involving spoken language, one invariably relies on relevantly annotated speech data for analysis and incorporation into such systems. The authors investigate methods and parameters for a baseline phonetic segmentation system on a few South African languages with the intention of determining how accurately they can apply basic methods and characterising typical deficiencies with the goal of defining further refinement strategies. An HMM-based system with a single mixture per triphone is found to work well, though the accurate segmentation of plosives remains a challenge. Suggestions for addressing this challenge are presentedenPhonetic segmentationHidden markov modelsTriphoneConference PresentationVan Niekerk, D., & Barnard, E. (2007). Important factors in HMM-based phonetic segmentation. 18th Annual Symposium of the Pattern Recognition Association of South Africa (PRASA). http://hdl.handle.net/10204/1978Van Niekerk, DR, and E Barnard. "Important factors in HMM-based phonetic segmentation." (2007): http://hdl.handle.net/10204/1978Van Niekerk D, Barnard E, Important factors in HMM-based phonetic segmentation; 18th Annual Symposium of the Pattern Recognition Association of South Africa (PRASA); 2007. http://hdl.handle.net/10204/1978 .TY - Conference Presentation AU - Van Niekerk, DR AU - Barnard, E AB - When doing research into or building systems involving spoken language, one invariably relies on relevantly annotated speech data for analysis and incorporation into such systems. The authors investigate methods and parameters for a baseline phonetic segmentation system on a few South African languages with the intention of determining how accurately they can apply basic methods and characterising typical deficiencies with the goal of defining further refinement strategies. An HMM-based system with a single mixture per triphone is found to work well, though the accurate segmentation of plosives remains a challenge. Suggestions for addressing this challenge are presented DA - 2007-11 DB - ResearchSpace DP - CSIR KW - Phonetic segmentation KW - Hidden markov models KW - Triphone LK - https://researchspace.csir.co.za PY - 2007 SM - 978-1-86840-656-2 T1 - Important factors in HMM-based phonetic segmentation TI - Important factors in HMM-based phonetic segmentation UR - http://hdl.handle.net/10204/1978 ER -