Cepstral Domain Segmental Nonlinear Feature Transformations for Robust Speech Recognition | Publicación