Prosody evaluation for embedded slovene speech-synthesis systems

This paper describes an evaluation of the prosody modeling in an HMM-based Slovene speech-synthesis system that is suitable for embedded systems due to its relatively small memory footprint. The objective-evaluation procedure is based on the results of the automatic recognition of syntactic-prosodic boundary positions and accented words in the synthetic speech. We have shown that the recognition results represent a close match with the prosodic notations, labeled by the human expert on the natural-speech counterpart produced by the speaker whose speech was used to train the speech-synthesis system. Therefore, the recognition rate of the prosodic events is proposed as an objective evaluation measure for the quality of the prosodic modeling in the speech-synthesis system. The results of the proposed evaluation method are also in accordance with previous subjective-listening evaluation tests, where high scores for the naturalness for such a type of speech synthesis were observed.