An objective measure for assessment of a corpus-based text-to-speech system

The primary goal of this work is to find an objective measure to track the performance of a corpus-based TTS system. This paper proposed a new objective measure using the dynamic time warping (DTW) algorithm to calculate the distortion distance between the original utterances originated from the speech corpus and the target utterances synthesized from the text corpus. A formal subjective study shows that the new objective scores are consistency with the subjective results.