The 1st Workshop on Child, Computer and Interaction (WOCCI2008)

Chania, Crete, Greece
October 23, 2008

Recent Advances In Sonic Italian Children’s Speech Recognition For Interactive Literacy Tutors

Piero Cosi

Istituto di Scienze e Tecnologie della Cognizione, C.N.R., Padova, Italy

Recent advances in SONIC Italian children’s speech recognition will be described. This work, completing a previous one developed in the past, was conducted with the specific goals of integrating the newly trained children’s speech recognition models into the Italian version of the Colorado Literacy Tutor platform. Specifically, children’s speech recognition research for Italian was conducted using the complete training and test set of the ITC-irst Children’s Speech Corpus. Using the University of Colorado SONIC LVSR system, we demonstrate a phonetic recognition error rate of 12,0% for a system which incorporates Vocal Tract Length Normalization (VTLN), Speaker-Adaptive Trained phonetic models, as well as unsupervised Structural MAP Linear Regression (SMAPLR).

Full Paper

Bibliographic reference.  Cosi, Piero (2008): "Recent advances in sonic Italian children²s speech recognition for interactive literacy tutors", In WOCCI-2008, paper 02.