Hermitian-Based Hidden Activation Functions for Adaptation of Hybrid HMM/ANN Models

This work is concerned with speaker adaptation techniques for

artificial neural network (ANN) implemented as feed-forward

multi-layer perceptrons (MLPs) in the context of large vocabulary

continuous speech recognition (LVCSR). Most successful

speaker adaptation techniques for MLPs consist of augmenting

the neural architecture with a linear transformation network

connected to either the input or the output layer. The weights

of this additional linear layer are learned during the adaptation

phase while all of the other weights are kept frozen in order

to avoid over-fitting. In doing so, the structure of the speakerdependent

(SD) and speaker-independent (SI) architecture differs

and the number of adaptation parameters depends upon the

dimension of either the input or output layers. We propose an

alternative neural architecture for speaker-adaptation to overcome

the limits of current approaches. This neural architecture

adopts hidden activation functions that can be learned directly

from the adaptation data. This adaptive capability of the hidden

activation function is achieved through the use of orthonormal

Hermite polynomials. Experimental evidence gathered on the

Wall Street Journal Nov92 task demonstrates the viability of

the proposed technique.

annadapt.pdf
PDF file

In  Interspeech

Details

TypeInproceedings
> Publications > Hermitian-Based Hidden Activation Functions for Adaptation of Hybrid HMM/ANN Models