• Phonetics and Speech Sciences
• /
• v.7 no.4
• /
• pp.41-47
• /
• 2015

This paper examines whether the Global Vector model is applicable to Korean data as a universal learning algorithm. The main purpose of this study is to compare the global vector model (GloVe) with the word2vec models such as a continuous bag-of-words (CBOW) model and a skip-gram (SG) model. For this purpose, we conducted an experiment by employing an evaluation corpus consisting of 70 target words and 819 pairs of Korean words for word similarities and analogies, respectively. Results of the word similarity task indicated that the Pearson correlation coefficients of 0.3133 as compared with the human judgement in GloVe, 0.2637 in CBOW and 0.2177 in SG. The word analogy task showed that the overall accuracy rate of 67% in semantic and syntactic relations was obtained in GloVe, 66% in CBOW and 57% in SG.

• Speech Sciences
• /
• v.6
• /
• pp.25-32
• /
• 1999

Not only Korean scholars but also many scholars in the field of phonetic science were interested in Korean forced sounds from the viewpoint of general phonetics as a way to clarify the process of production and the physical characteristics of speech sounds. The author also tried to elucidate the characteristics of Korean forced sounds using the sound spectrograph (Umeda & Umeda 1965). Over the past 30 years since this study, many scholars have analyzed these sounds from various standpoints resulting in copious amounts of literature. In this paper, the author critically examined the results of previous studies on this subject dealing with VOT, spectral characteristics, fundamental frequency, states of glottis obtained from fiber-optic investigations, and patterns of tongue-palate contact acquired from observation by use of dynamic palatograph. These points were discussed in relation to the author's field of investigation.

• Phonetics and Speech Sciences
• /
• v.7 no.1
• /
• pp.11-17
• /
• 2015

This paper deals with the correlation between physical fatigue and speech signals. A treadmill task to increase fatigue and a set of subjective questionnaire for rating tiredness were designed. The results from the questionnaire and the collected bio-signals showed that the designed task imposes physical fatigue. The t-test for two-related-samples between the speech signals and fatigue showed that the parameters statistically significant to fatigue are fundamental frequency, first and second formant frequencies, long term average spectral slope, smoothed pitch perturbation quotient, relative average perturbation, pitch perturbation quotient, cepstral peak prominence, and harmonics to noise ratio. According to the experimental results, it is shown that mouth is opened small and voice is changed to be breathy as the physical fatigue accumulates.

• Phonetics and Speech Sciences
• /
• v.7 no.1
• /
• pp.125-131
• /
• 2015

The current study explores whether speaker variability in the fronted articulations of Seoul Korean affricates can be explained by cognitive differences measured by individual autistic traits. The goal was to explore Yu's (2010; 2013) proposal that individual differences in cognitive style can be an important factor in speakers' use of sound variants. The spectral peak frequencies (SPF) of affricates relative to those of fricatives, reported in Kong et al. (2014), were used to acoustically represent the relative degree of anterior place of constriction. When these individual SPFs were related to the scores of Autistic-Spectrum Quotients (Baron-Cohen et al., 2001), a correlation was found for the male speakers, but not for the female speakers, such that speakers of more anterior affricate productions scored low in AQs. Discussion is made with respect to how these findings are in line with Yu's proposal.

• Phonetics and Speech Sciences
• /
• v.7 no.1
• /
• pp.133-138
• /
• 2015

The purpose of this study is to provide deeper understanding of how it is possible for people to understand PC communication language they have never seen or heard before without any problem. In order to answer this question, we focused on the vowel variation through which new variants are created (for PC communication), and hypothesized that there is a phonetic constraint which requires the vowel of the variant to be phonetically similar (to the maximum) to the vowel of the original word. Through the corpus analysis of the dictionary of PC communication language, we show that our hypothesis is justified by the fact that most of the variants we collected from the dictionary, that is, 90% of them, conformed to the phonetic constraint we postulated.

• Phonetics and Speech Sciences
• /
• v.7 no.1
• /
• pp.47-55
• /
• 2015

This study examined whether speech acceptability was changed under various conditions of prosodic manipulations. Both speech rate and voice loudness reportedly are associated with acceptability and intelligibility. Speech samples by twelve speakers with mild dysarthria were recorded. Speech rate and loudness changes were made by digitally manipulating habitual sentences. 3 different loudness levels (70, 75, & 80dB) and 4 different speech rates (normal, 20% rapidly, 20% slowly, & 40% slowly) were presented to 12 SLPs (speech language pathologists). SLPs evaluated sentence acceptability by 7-point Likert scale. Repeated ANOVA were conducted to determine if the prosodic type of resynthesized cue resulted in a significant change in speech acceptability. A faster speech rate (20% rapidly) rather than habitual and slower rates (20%, 40% slowly) resulted in significant improvement in acceptability ratings (p <.001). An increased vocal loudness (up to 80dB) resulted in significant improvement in acceptability ratings (p <.05). Speech rate and loudness changes in the prosodic properties of speech may contribute to improved acceptability.

• Phonetics and Speech Sciences
• /
• v.5 no.2
• /
• pp.81-90
• /
• 2013

The purposes of this study are to: (1) Compare the vowel production of English front vowels produced by Korean speakers using regional dialects and; (2) Investigate and compare the effectiveness of pronunciation training for each regional dialect group. To test these objectives, the English front vowels produced by five Youngnam dialect male speakers, five Youngnam dialect female speakers, five Kangwon dialect male speakers, and five Kangwon dialect female speakers were scrutinized. These dialect groups' vowel formants and length of English front vowels were evaluated, and the post-pronunciation training values were compared with those of pre-training values. The results indicate that pronunciation training is more effective for Youngnam dialect speakers, whilst both dialect groups have more success mastering the pronunciation of /${\varepsilon}$/ over /${\ae}$/.

• Phonetics and Speech Sciences
• /
• v.5 no.2
• /
• pp.53-58
• /
• 2013

The purpose of this study is to compare the first two vowel formants of the stressed and unstressed English vowels produced by ten young males (in their twenties and thirties) and ten old males (in their forties or fifties) from the Buckeye Corpus of Conversational Speech. The results indicate that the stressed and unstressed vowels, /i/ and $/{\ae}/$ in particular, from the two groups are different in their formant frequencies. In addition, the vowel space of the unstressed vowels is somewhat smaller than that of the stressed vowels. Specifically, the range of the second formant of the unstressed vowels and that of the first formant of the unstressed front vowels were compressed. The findings from this study can be applied to the pronunciation training for the Korean learners of English vowels. We propose that teachers of English pay attention to the stress patterns of English vowels as well as their formant frequencies.

• Phonetics and Speech Sciences
• /
• v.5 no.4
• /
• pp.145-155
• /
• 2013

This study investigated the articulatory characteristics of Korean singleton and geminate sonorants in the word-medial position, effects of the duration of the sonorant consonant and the preceding vowel on perception, and the difference between native Korean speakers and foreign learners of Korean in perceiving the singleton and geminate consonant contrast. The Korean sonorant consonants(/m, n, l/) are examined from the VCCV, VCV sequences through speech production and perception experiments. The results suggest that the duration of the sonorant consonant is the most important factor for native Korean speakers to recognize whether sonorants are overlapped, and the duration of preceding vowel and other factors affect the recognition of singleton/geminate consonant contrast if the duration is not obvious. A perception experiment showed Chinese Korean language learners did not clearly distinguish singleton consonants from geminate consonants. The results of this study provide basic data for recognition of singleton/geminate consonant contrast in word-medial of Korean language, and can be utilized for teaching Korean pronunciation as a foreign language.

• Phonetics and Speech Sciences
• /
• v.5 no.4
• /
• pp.121-128
• /
• 2013

In Korean, wh-question and indefinite yes-no questions have the same morphemic and syntactic structure. In speech, however, these two types of questions are distinguished by a prosodic difference. In this study, we examined if Chinese Korean leaners can distinguish between these two types of questions in production and if they can correctly perceive the different meaning of a question based on the prosodic information. For this purpose, we analysed two types of interrogative sentences produced by 5 native speakers and 15 Chinese Korean language leaners. The results show that the 5 Korean native speakers produce two types of questions by a salient prosodic difference, i.e., difference of prosodic structure, different pitch range of wh-phrase and indefinite phrase, and different boundary tone. However, for the 15 Chinese speakers, the two types of questions were not distinguished by the same prosodic features but in the perception analysis they were able to distinguish between the two types of questions easily.