AUTOMATIC ACCENT CLASSIFICATION USING ARTIFICIAL NEURAL NETWORKS
C.S. Blackburn, J.P. Vonwiller and R.W. King
1993
We describe the development and performance of an automatic English accent classification system to discriminate between the speech of subjects whose first language is Arabic, Chinese and Australian English. The system operates on speech samples of arbitrary duration. The classification is performed in stages. A broad phonetic class segmenter divides incoming speech into one of voiced, unvoiced, stop and energy dip. For each of these segment types an artificial neural network is used to classify the accent. The sequence of accent labels from these four networks is examined to obtain a cumulative measure of the accent classification. Tested on a small set of data the system has been found to correctly classify accents as rapidly as a trained phonetician.
If you have difficulty viewing files that end '.gz',
which are gzip compressed, then you may be able to find
tools to uncompress them at the gzip
web site.
If you have difficulty viewing files that are in PostScript, (ending
'.ps' or '.ps.gz'), then you may be able to
find tools to view them at
the gsview
web site.
We have attempted to provide automatically generated PDF copies of documents for which only PostScript versions have previously been available. These are clearly marked in the database - due to the nature of the automatic conversion process, they are likely to be badly aliased when viewed at default resolution on screen by acroread.