Some Properties of Autoregressive Model Related Cepstrum

Autoregressive model related cepstrum, or briefly LPC cepstrum, recently gained renewed attention because it was shown [1-3] to be an effective representation in speech recognizer designs. In this paper, we investigate some properties of the LPC cepstrum that are of important considerations when the recognizer is to be deployed in situations where mismatch between the training conditions and the testing conditions may potentially occur.