Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network

  Advanced Search  

Adaptation of childrens speech with limited data based on formant-like peak alignment

Summary: Adaptation of childrenŐs speech with limited data based
on formant-like peak alignment
Xiaodong Cui *, Abeer Alwan
Department of Electrical Engineering, The Henry Samuli School of Engineering and Applied Science, 66-147E
Engr. IV, 405 Hilgard Avenue, Box 951594, University of California, Los Angeles, CA 90095-1594, USA
Received 24 April 2004; received in revised form 21 April 2005; accepted 13 May 2005
Available online 12 July 2005
Automatic recognition of childrenŐs speech using acoustic models trained by adults results in poor per-
formance due to differences in speech acoustics. These acoustical differences are a consequence of children
having shorter vocal tracts and smaller vocal cords than adults. Hence, speaker adaptation needs to be per-
formed. However, in real-world applications, the amount of adaptation data available may be less than
what is needed by common speaker adaptation techniques to yield reasonable performance. In this paper,
we first study, in the discrete frequency domain, the relationship between frequency warping in the front-
end and corresponding transformations in the back-end. Three common feature extraction schemes are
investigated and their transformation linearity in the back-end are discussed. In particular, we show that
under certain approximations, frequency warping of MFCC features with Mel-warped triangular filter
banks equals a linear transformation in the cepstral space. Based on that linear transformation, a for-
mant-like peak alignment algorithm is proposed to adapt adult acoustic models to childrenŐs speech.
The peaks are estimated by Gaussian mixtures using the Expectation-Maximization (EM) algorithm [Zol-


Source: Alwan, Abeer - Electrical Engineering Department, University of California at Los Angeles


Collections: Computer Technologies and Information Sciences