5th European Conference on Speech Communication and Technology

Rhodes, Greece
September 22-25, 1997

Automatic Speech Recognition for Children

Alexandros Potamianos, Shrikanth Narayanan, Sungbok Lee

AT&T Labs-Research, Florham Park, NJ, USA

In this paper, the acoustic and linguistic characteristics of children speech are investigated in the context of automatic speech recognition. Acoustic variability is identified as a major hurdle in building high performance ASR applications for children. A simple speaker normalization algorithm combining frequency warping and spectral shaping introduced in [5] is shown to reduce acoustic variability and significantly improve recognition performance for children speakers (by 25{ 45%). Age-dependent acoustic modeling further reduces word error rate by 10%. Piece-wise linear and phoneme-dependent frequency warping algorithms are proposed for reducing acoustic mismatch between the children and adult acoustic spaces.

Full Paper

Bibliographic reference.  Potamianos, Alexandros / Narayanan, Shrikanth / Lee, Sungbok (1997): "Automatic speech recognition for children", In EUROSPEECH-1997, 2371-2374.