Multi-resolution phonetic/segmental features and models for HMM-based speech recognition.

Saeed Vaseghi, Naomi Harte, Ben Milner

Browse the full ICASSP paper archive.