In this paper, the prosody of a parallel multispeaking rate Mandarin read speech corpus is investigated. The corpus contains four parallel speech datasets uttered by a female professional announcer with various speech rates (SRs) of 4.40 (fast), 3.82 (nor
