EUROSPEECH 2001 Scandinavia
In this paper we present an approach to introducing more phonetically motivated information into automatic speech recognition in the form of a phonetic `expert'. To avoid the curse of dimensionality problem, the expert information is introduced at the level of the acoustic model. Two types of experts are used each providing discriminative information regarding groups of phonetically related phonemes. The phonetic expert is implemented using an MLP. A numbers recognition task shows that, when using the expert in conjunction with both a fullband and a multiband system speech recognition performances are increased.
Bibliographic reference. Christensenyz, Heidi / Lindbergy, BÝrge / Anderseny, Ove (2001): "Introducing phonetically motivated information into ASR", In EUROSPEECH-2001, 2289-2292.