The words contained in this file might help you see if this file matches what you are looking for:

...Creation of a speech to text system for kiswahili dr katherine getao and evans miriti university nairobi abstract the any language is an onerous task this especially case if not much has been done in area processing that previously lack or little availability previous work means there are few resources on which one can build upon be considered consist four main components these front end emission model lexicon finally component combines three perform transcription from acoustic signal paper we will describe implementation training also viterbi algorithm chose use our limitations approach give suggestions improvement introduction dictation described as extended monologue by single speaker systems fall within realm large vocabulary continuous recognition lvcsr they have over words jurafsky martin several now available market some provide rates per minute word accuracy about vocabularies up http www dragontalk com natural htm two most commonly used ibm s viavoice pro usb edition with drag...