• Speech Sciences
• /
• v.7 no.2
• /
• pp.209-224
• /
• 2000

This paper proposes an automatic detection method of intonational and accentual phrases in Korean standard continuous speech. We use the pause over 150 msec for detecting intonational phrases, and extract accentual phrases from the intonational phrases by analyzing syllables and pitch contours. The speech data for the experiment are composed of seven male voices and two female voices which read the texts of the fable 'the ant and the grasshopper' and a newspaper article 'manmulsang' in normal speed and in Korean standard variation. The results of the experiment shows that the detection rate of intonational phrases is 95% on the average and that of accentual phrases is 73%. This detection rate implies that we can segment the continuous speech into smaller units(i.e. prosodic phrases) by using the prosodic information and so the objects of speech recognition can narrow down to words or phrases in continuous speech.

• Speech Sciences
• /
• v.15 no.4
• /
• pp.67-81
• /
• 2008

The accuracy rate of P2G (Phoneme-to-Grapheme) is one of the important factors determining the quality of unlimited voice recognition (VR) systems. Few studies were, however, conducted to reduce ambiguities of a phoneme string which can be segmented into a variety of different linguistic units (i.e. morphemes, words, eo-jeols), thus be transformed into more than one grapheme string. This paper is a preliminary research for building a large knowledge base of those homonymic & heterographic units(HHUs), which will provide unlimited Korean VR systems with more accurate P2G information. This paper analyzes 2 main factors generating HHUs: (1) boundary determination of the prosodic unit; (2) its segmentation into linguistic units. In this paper, linguistic characteristics determining variable boundaries of a prosodic unit are investigated, and the ambiguity types of HHUs are classified in accordance with their morphological and syntactic structures as well as with the phonological rules governing them.

• 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.