Processing spoken lectures in resource-scarce environments
Loading...
Date
Authors
van Heerden, Charl
De Villiers, Pieter
Barnard, Etienne
Davel, Marelie H.
Journal Title
Journal ISSN
Volume Title
Publisher
Pattern Recognition Association of South Africa and Mechatronics International Conference
Abstract
Initial work towards processing Afrikaans spoken
lectures in a resource-scarce environment is presented. Two
approaches to acoustic modeling for eventual alignment are
compared: (a) using a well-trained target-language acoustic
model and (b) using an acoustic model from another language, in
this case American English. We show that while target-language
acoustic models are preferable, similar performance can be
achieved by repeatedly bootstrapping with the American English
model, segmenting and then adapting or training new models
using the segmented spoken lectures. The eventual systems
perform quite well, aligning more than 90% of a selected set
of target words successfully.
Description
Citation
Charl J van Heerden, Pieter de Villiers, Etienne Barnard and Marelie H Davel, “Processing spoken lectures in resource-scarce environments”, in Proc. Annual Symp. Pattern Recognition Association of South Africa (PRASA), pp 138-143, Vanderbijlpark, South Africa, 2011. [http://engineering.nwu.ac.za/multilingual-speech-technologies-must/publications]