• Speech Sciences
• /
• v.10 no.4
• /
• pp.7-21
• /
• 2003

Sums-of-Products models were built for segment duration prediction of spoken Korean. An experiment for the modelling was carried out to apply the results to Korean text-to-speech synthesis systems. 670 read sentences were analyzed. trained and tested for the construction of the duration models. Traditional sequential rule systems were extended to simple additive, multiplicative and additive-multiplicative models based on Sums-of-Products modelling. The parameters used in the modelling include the properties of the target segment and its neighbors and the target segment's position in the prosodic structure. Two optimisation strategies were used: the downhill simplex method and the simulated annealing method. The performance of the models was measured by the correlation coefficient and the root mean squared prediction error (RMSE) between actual and predicted duration in the test data. The best performance was obtained when the data was trained and tested by ' additive-multiplicative models. ' The correlation for the vowel duration prediction was 0.69 and the RMSE. 31.80 ms. while the correlation for the consonant duration prediction was 0.54 and the RMSE. 29.02 ms. The results were not good enough to be applied to the real-time text-to-speech systems. Further investigation of feature interactions is required for the better performance of the Sums-of-Products models.

• Korean Journal of Ichthyology
• /
• v.18 no.1
• /
• pp.45-54
• /
• 2006

Age and growth of the small yellow croaker, Larimichthys polyactis were estimated using right sagittal otoliths of 506 fish specimens from March to December, 2002 and from January to February, 2005 in the South Sea, part of the East China Sea of Korea. Examination of outer margins of the otolith showed that the opaque zone was formed once a year. Marginal increment of the otolith formed annual rings from May and June at the beginning of spawning season. In the relationship between total length and body weight, a multiplicative error structure was assumed because variability in growth increased as a function of the length, and the estimated equation was $BW=0.0044TL^{3.2502}$ ($R^2=0.97$). The relative growth as body weight at total length has significant difference between females and males (P<0.05). For describing growth of the small yellow croaker, Larimichthys polyactis a von Bertalanffy growth model was adopted. The von Bertalanffy growth curve had an additive error structure and the growth parameters estimated from non-linear regression were $L_{\infty}=33.88cm$, K=0.20/year and $t_0=-2.39year$. Growth at age of males and females shows no significant difference (P>0.05). Most examined fish were 1, 2 and 3 years old, although the oldest fish were 7 old for males and 8 for females.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.36 no.3
• /
• pp.222-233
• /
• 2000

A methodology for determining age by otolith of small yellow croaker, Pseudosoiaena polyactis Bleeker, was developed. A thin section method was chosen to be suitable for age determination because the otolith had a three-dimensional shape and thus it was not possible to read the otolith rings on the surface. The clear rings were identified on the vertical-axis cross-sectioned otoliths. The total length-total weight relationship and the growth parameters were estimated with error structure to endow with accuracy. In the relationship between total length and total weight, a multiplicative error structure was assumed because variability in growth increased as a function of the length, and the estimated equation was $W=0.0049L^{3.2153}$. The variability in growth was constant as a function of the age, revealing an additive error structure. The von Bertalanffy growth parameters were obtained from a nonlinear regression as $L_{\infty}= 37.11cm, K=0.20/yr and t_0=-1.88.$

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.44 no.4
• /
• pp.312-322
• /
• 2008

The age and growth of the sandfish, Arctoscopus japonicus were investigated from samples of the eastern sea danish seine and gill net fishery in the East Sea of Korea from February, 2004 to December, 2007. Ages were determined from annuli in otoliths and annuli were formed between December and February once a year. Also, the main spawning period was estimated to be between December and January, thus rings were considered to be annual marks. For the relationship between fork length and total weight, a multiplicative error structure was assumed because variability in growth increased as a function of the length. The relationship between fork length and total weight were $TW=0.0083FL^{3.01}(R^2=0.966)$ for female and $TW=0.0079FL^{3.04}(R^2=0.969)$ for male. The variability in growth was constant as a function of age, revealing an additive error structure. The von Bertalanffy growth parameters estimated by a non-linear method were $L_{\infty}=31.6cm$, K=0.266/year, $t_0=-0.835$ years for female and $L_{\infty}=27.7cm$, K=0.273/year, $t_0=-0.806$ years for male. Growth at age between male and female had a significant difference(P<0.001).

• Korean Journal of Ichthyology
• /
• v.19 no.4
• /
• pp.324-331
• /
• 2007

Age and growth of the robust tonguefish, Cynoglossus robustus were estimated using scale of 353 fish specimens from February, 2004 to December, 2005 in the Southern Sea of Korea. Marginal increment of the scale formed annual rings from October to November at the beginning of autumn season. In the relationship between total length and body weight, a multiplicative error structure was assumed because variability in growth increased as a function of the length, and the estimated equation was $BW=0.0013TL^{3.399}$ ($R^2=0.916$). The relative growth as body weight at total length has significant difference between females and males (p<0.05). For describing growth of the robust tonguefish, C. robustus a von Bertalanffy growth model was adopted. The von Betalanffy growth curve had a additive error structure and the growth parameters estimated from Walford method were $L_{\infty}=43.77cm$, K=0.186/year and $t_0=-2.295year$. Growth at age of females and males shows no significant difference (P>0.05).