• Phonetics and Speech Sciences
• /
• v.10 no.2
• /
• pp.45-50
• /
• 2018

The vowel/consonant ratio is a well-known cue for the voicing of postvocalic consonants. This study investigates how this ratio changes as a function of speaking rate. Seven speakers of North American English read sentences containing target monosyllabic words that contrasted in coda voicing at three different speaking rates. Duration measures were taken for the voice onset time (VOT) of the onset consonant, the vowel, and the coda. The results show that the durations of the onset VOT and vowel are longer before voiced codas, and that the durations of all segments increase monotonically as speaking rate decreases. Importantly, the vowel/consonant ratio, a primary acoustic cue for coda voicing, was found to pattern asymmetrically for voiced and voiceless codas; it increases for voiced codas but decreases for voiceless codas with the decrease in speaking rate. This finding suggests that there is no stable ratio in the duration of preconsonantal vowels that is maintained in different speaking styles.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1994.06a
• /
• pp.997-1002
• /
• 1994

Predictive neural network models are powerful speech recognition models based on a nonlinear pattern prediction. Those models can effectively normalize the temporal and spatial variability of speech signals. But those models suffer from poor discrimination between acoustically similar words. In this paper, we propose a discriminative training algorithm for predictive neural network models based on a generalized probabilistic descent (GPD) algorithm and minimum classification error formulation (MCEF). The Evaluation of our training algorithm on ten Korean digits shows its effectiveness by 40% reduction of recognition error.

• Speech Sciences
• /
• v.10 no.4
• /
• pp.63-71
• /
• 2003

This study aims to experiment on learning French intonation, based on the visualization of melody, which was employed in the early sixties to reeducate those with communication disorders. The visualization of melody in this paper, however, was used to the foreign language learning and produced successful results in many ways, especially in learning foreign intonation. In this paper, we used the PitchWorks to visualize some French intonation samples and experiment on learning intonation based on the bitmap picture projected on a screen. The students could see the melody curve while listening to the sentences. We could observe great achievement on the part of the students in learning intonations, as verified by the result of this experiment. The students were much more motivated in learning and showed greater improvement in recognizing intonation contour than just learning by hearing. But lack of animation in the bitmap file could make the experiment nothing but a boring pattern practices. It would be better if we can use a sound analyser, as like for instance a PitchWorks, which is designed to analyse the pitch, since the students can actually see their own fluctuating intonation visualized on the screen.

• Phonetics and Speech Sciences
• /
• v.1 no.1
• /
• pp.55-62
• /
• 2009

This paper proposes an improved method to verify a keyword segment that results from a word spotting system. First a baseline word spotting system is implemented. In order to improve performance of the word spotting systems, we use a two-pass structure which consists of a word spotting system and an utterance verification system. Using the basic likelihood ratio test (LRT) based utterance verification system to verify the keywords, there have been certain problems which lead to performance degradation. So, we propose a method which uses phone-level log-likelihood ratios (PLLR) patterns in computing confidence measures for each keyword. The proposed method generates weights according to the PLLR patterns and assigns different weights to each phone in the process of generating confidence measures for the keywords. This proposed method has shown to be more appropriate to word spotting systems and we can achieve improvement in final word spotting accuracy.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.2
• /
• pp.99-107
• /
• 2018

This study started from the public interest purpose to help accessibility for the information acquisition of the vulnerable groups due to visual difficulties such as the elderly and the visually impaired. In this study, the server resources are minimized and implemented in most of the user smart phones. In addition, we implement a method to gather necessary information by collecting only pattern information by utilizing crawl & clipping without having to visit the site of the information of the various sites having the data necessary for the user, and to have it in the server. Especially, we applied the TTS(Text-To-Speech) service composed of smart phone apps and tried to develop a unified customized information collection service based on voice-based information collection method.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.29B no.10
• /
• pp.90-99
• /
• 1992

In this paper, we propose a neural network model for speech recognition. The model consists of feature extraction parts and recognition parts. The interconnection model based on ${\Delta}^2$G operator was used for frequency analysis. Two features, global feature and local feature, were extracted from this model. Recognition parts consist of global grouping stage and local grouping stage. When the input pattern was coded by slope method, the recognition rate of speakers, A and B, was 100%. When the test was performed with the data of 9 speakers, the recognition rate of 91.4% was obtained.

• Speech Sciences
• /
• v.9 no.2
• /
• pp.13-23
• /
• 2002

This study examines the effect of laryngeal consonants of Korean on prosodic domain-initial strengthening. Keating, Cho, Fougeron & Hsu (1999), Fougeron & Keating (1996), and Hsu & Jun (1998) found that consonants at the beginnings of larger phrases are more constricted than consonants at the beginnings of smaller phrases. Korean laryngeal consonants pose a counter-example to the general pattern of domain-initial strengthening since tense and aspirated consonants are longer word-medially than word-initially. Previous work on domain-initial strengthening focused on domain-initial consonants at different prosodic domains. This study shows that acoustic cues that are not domain-edge also function to demarcate prosodic structure when the domain-initial consonant is laryngeal: VOT for an aspirated consonant and duration of V2 for a tense consonant.

• Speech Sciences
• /
• v.13 no.3
• /
• pp.121-132
• /
• 2006

In Hangeul, the graphemes of syllables are organized in horizontal, vertical and mixed (both horizontal & vertical) orientations, and the graphemic position of consonant(s) and vowel(s) within a each syllable needs to be maintained within a square pattern. We investigated the characteristics of writing errors of 9 stroke patients with right hemisphere (RH) lesions and compared it to the performances of 15 normal subjects. The subjects were asked to write to dictation of 90 Korean syllables. One of the interesting findings was that our patients manifested visuospatial errors which are not commonly observed in other language-speaking (e.g., English) patients due to the unique syllabic organizations of Korean writing system. The prominent errors in the RH group could be explained by the impaired RH which normally controls the visuospatial functions.

• Speech Sciences
• /
• v.14 no.2
• /
• pp.139-155
• /
• 2007

Old Korean regulated poems have a typical prosodic structure characterized by a pitch contour. This work applies Jun's finding in Seoul Korean(Jun 1993, 2000, 2005) to old Korean regulated poems, and reports some other significant phonetic characteristics, arguing that old Korean regulated poems have a regular rhythm based on the pitch contour implementing the typically hierarchical prosodic structure. The major prosodic units defined are a foot, a phrase, and a line. Next, this work proposes pitch contour characterizing prominence in a unit, boundary tones, and pauses at the boundary position, as the basic and significant cues of rhythm of a Korean poem. Specifically, some significant characteristics are discussed as follows: first, the tonal pattern of a foot is HL, starting high and ending low; second, the lowering boundary tones of HL% and L% are perceived at the end of a phrase and a line; and finally, a gradient degree of pause is observed at each unit-final position.

• Phonetics and Speech Sciences
• /
• v.3 no.1
• /
• pp.77-83
• /
• 2011

The present study reports the results of an acoustic study of nasal assimilation at word boundaries in Chinese-Korean interlanguage. Twelve Chinese learners of Korean and four Korean native speakers recorded obstruent#nasal sequences in noun compounds and verb phrases, and their different production patterns were examined in detail. While nasalization of the word-final obstruents occurred only in 11.7% of the obstruent#nasal sequences for the Chinese learners, the Korean native speakers showed complete nasalization of those sequences. However, there was small, but consistent effect of learning on the production of external sandhi in L2, because there were shown to be differences in the rate of nasalization between the two proficiency groups of Chinese participants. On average, the intermediate level learners nasalized the target stops at the rate of 16%, and the beginning level learners showed the 7% nasalization rate. In addition, it was found that the context difference such as noun compounds versus verb phrases does not influence the nasalization pattern across word boundaries.