EUROSPEECH 2001 Scandinavia
In this paper we develop a method of representing the speech waveform in terms of a set of abstract, linguistic distinctions in order to derive a set of discriminative features for use in a speech recognizer. By combining the distinctive feature representation with our original waveform representation we are able to achieve a reduction in word error rate of 33 percent on an automatic speech recognition task.
Bibliographic reference. Eide, Ellen (2001): "Distinctive features for use in an automatic speech recognition system", In EUROSPEECH-2001, 1613-1616.