Abstract
Speaker independent, spontaneous and continuous
speech recognition system (ASR) can be integrated to
other technologies like mobile to create an interface
between technology and illiterate people so that they
can use modern technologies. One of the major hurdles
in such ASR is unacceptable word error rate. The
paper explores the possibility of analyzing the Urdu
speech corpus based on recognition results to improve
word error rate (WER).
Saad Irtza, Sarmad Hussain , Sarmad Hussain . (2012) Error Analysis of Single Speaker Urdu Speech Recognition System, Conference on Language and Technology 2012.
-
Viewed
1475 -
Downloads
151
Publisher
Center for Language Engineering
Country
Pakistan
City
Lahore
From
09-11-2012
To
10-11-2012