Italian Children Speech Recognition with Application to Interactive Books and Tutors

This work represents a joint collaboration between the Center for Spoken Language Research (CSLR) at the University of Colorado and the Institute of Cognitive Sciences and Technologies of the National Research Council located in Padova Italy. This work was conducted with the specific goals of developing improved recognition of children's speech in Italian and the installation and integration of the children's speech recognition models into the Italian Literacy Tutor system. Specifically, children's speech recognition research for Italian was conducted using the ITC-irst Children's Speech Corpus (Giuliani & Gerosa, 2003). Using the University of Colorado SONIC large vocabulary speech recognition system, we demonstrate a phonetic recognition error rate of 13.5% for a system which incorporates Vocal Tract Length Normalization (VTLN), Cepstral variance normalization, Speaker-Adaptive Trained phonetic models, as well as iterative unsupervised Structural MAP Linear Regression (SMAPLR). These new acoustic models have been incorporated within an Italian version of the Colorado Literacy Tutor system.