• The Journal of the Acoustical Society of Korea
• /
• v.20 no.3
• /
• pp.24-33
• /
• 2001

this paper describes a baseline for an implementation of a corpus-based Korean TTS system. The conventional TTS systems using small-sized speech still generate machine-like synthetic speech. To overcome this problem we introduce the corpus-based TTS system which enables to generate natural synthetic speech without prosodic modifications. The corpus should be composed of a natural prosody of source speech and multiple instances of synthesis units. To make a phone level synthesis unit, we train a speech recognizer with the target speech, and then perform an automatic phoneme segmentation. We also detect the fine pitch period using Laryngo graph signals, which is used for prosodic feature extraction. For break strength allocation, 4 levels of break indices are decided as pause length and also attached to phones to reflect prosodic variations in phrase boundaries. To predict the break strength on texts, we utilize the statistical information of POS (Part-of-Speech) sequences. The best triphone sequences are selected by Viterbi search considering the minimization of accumulative Euclidean distance of concatenating distortion. To get high quality synthesis speech applicable to commercial purpose, we introduce a domain specific database. By adding domain specific database to general domain database, we can greatly improve the quality of synthetic speech on specific domain. From the subjective evaluation, the new Korean corpus-based TTS system shows better naturalness than the conventional demisyllable-based one.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.6
• /
• pp.84-88
• /
• 1999

The accurate estimation of segmental duration is crucial for natural-sounding text-to-speech synthesis. For predicting Korean segmental durations, conventional methods utilized phonemic context, part-of-speech context and locational information in prosodic phrase. In this paper, the tonal information of phonemes is employed for more accurate prediction. After defining two non-boundary tones and six boundary tones, we annotated the tonal label on each syllable of 400 sentences. To predict segmental duration using tonal information, we constructed neural networks with a real-valued output node predicting phonemic duration and trained them by backpropagation algorithm. Experimental results showed that the proposed features are effective for predicting Korean segmental durations, and we got 0.863 correlation coefficient of the observed durations and predicted ones.