• Proceedings of the Acoustical Society of Korea Conference
• /
• 1994.06a
• /
• pp.1021-1026
• /
• 1994

This paper proposes a filter structure suitable for speech synthesis applications. We first derive the lossy pole-zero model by employing the wave digital filter(WDF) adaptor formula, and by converting the fixed termination value - 1 into a loss factor $\mu$c$\in$(-1, 1). Then we discuss how to determine the reflection We employ the Durbin's method in estimating the numerator polynomial of the lossy pole-zero transfer function from the given speech sound, and then apply the step-down algorithm on the numerator to extract the reflection coefficients of the closed-termination tract. For determining the reflection coefficients of the other parts we employ a pre-calculated pole-estimator polynomial.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.1
• /
• pp.1-10
• /
• 2012

Improving operating frequency of processors is no longer today's issues; a multiprocessor technique which integrates many processors has received increasing attention. Currently, high-performance processors that integrate 64 or 128 cores are developing for large data processing over 2, 4, or 8 processor cores. This paper proposes an optimal many-core processor for synthesizing guitar sounds. Unlike the previous research in which a processing element (PE) was assigned to support one of guitar strings, this paper evaluates the impacts of mapping different numbers of PEs to one guitar string in terms of performance and both area and energy efficiencies using architectural and workload simulations. Experimental results show that the maximum area energy efficiencies were achieved at PEs=24 and 96, respectively, for synthesizing guitar sounds with sampling rate of 44.1kHz and 16-bit quantization. The synthesized sounds were very similar to original guitar sounds in their spectra. In addition, the proposed many-core processor was 1,235 and 22 times better than TI TMS320C6416 in area and energy efficiencies, respectively.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.13 no.4
• /
• pp.145-149
• /
• 1980

Various methods have been so far proposed to find out the directions and spectra of sound waves from the sources for provisions of noise controls. The conventional methods are generally classified into three systems such as, single microphone system, moving microphone system and multi-microphone system, which composes a resultant super directivity by giving a appropriate delay and a weighting coefficient in the output of each microphone. In case of using a single microphone there is a difficulty in providing it with desirable super directivity in the low frequency range, while in case of using multi-microphone system there has been a disadvantage that the measurement of directivity could not separately be done with the spectrum analysing. And in case of the use of a moving microphone system it needs a condition that the sound source to be detected should be stationary state and in rest. However here we introduce a method that the spectral analysing and the directivity of synthesis can be separately carried out by using a linear array of many microphones, in which each output of the microphone is multiplied by appropriate weighting coefficient and all of those products are summed after passing through adequate filters. The resultant signal is then sampled with an adequate sampling frequency and taken average for processing.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.38 no.5
• /
• pp.44-57
• /
• 2001

In order to synthesize grammatically correct Korean voices, we have to refer to the 'Standard Pronunciation Rules(SPR)' stated in the 'Standard Grammar of Korean Language.' Therefore, the rules that is used for a Korean-voice-synthesis system to find Korean voices corresponding to a given Korean sentence must completely reflect the SPR and must be sound. However, in the field of computer science they have just used the SPR without proving the completeness and soundness of their rules. In this paper, we construct a Petri net model for each rule of SPR, integrate all the Petri net models to build one big Petri net completely representing SPR, and analyse the Petri net to prove the consistency of it. Then, we transfer the Petri net model into a look-up table for Korean voice. Using this table, we can avoid the drawbacks of existing approaches such as going through several stages or repetitively applying a converting process.

• Journal of Korean Classical Literature and Education
• /
• no.15
• /
• pp.329-353
• /
• 2008

Tasan Jeong Yak-yong is great scholar, who makes a synthesis of Sil-hak[實學, Practical Science of Korea], reformer of society, and a poet in the Joseon Dynasty. He expressed contradiction and conflict of those days by intellectual language, and reperceived basic ideology of the Joseon society. Also he theorized dissatisfaction of the people about those days and its system as form of religion. We can divide Tasan's life into two times. The first part is his ages 16~39 in the period of Jeong-jo(1777~1800). The second part is in the period of Sun-jo(1801~1834). In this period, he was exiled into Gang-jin for 17 years. After banishment, he lived a quiet life for the rest of his life in his hometown. His allegoric poetry were written in this second period. The special feature of allegoric poetry is strong satire. An allegory would be that is 'king's ear', which the barber has sight, or the barber's voice, which has divulged king's secret among the bamboos. Otherwise it would be that is the sound 'king's ear is donkey's ear' in the bamboos. This sound is divulging of the true donkey's ear. It doesn't travel to audiences, but travels trough wind in the bamboos. The narration exists just as story that barber can't stand to keep silence about king's secret. There are exposure of true and critical motive as allegoric expression. Tasan's allegoric poetry stand on the basis of his love for the people. Also there reveals his thought deeply with an enormous amount of reading and self-communion. Moreover there are his warm mind with his sharp insight in which captures alive lives as allegoric materials. Most of allegoric poetry satirize actuality of those days to make an excuse for external distinguishing marks of animals and plants. However Tasan's poetry are different from them. After he grasped serious problems from his contemporary actuality, and then choosed allegoric media to express correctly. Because he grasped the special features of lives after minute observation, he could exposure controversial point of the actual. His sharp insight was not limited to allegoric media. He noticed his period and the current of his society sensitively. It made his allegoric poetry as important materials to make us to know the condition of the people in the Joseon Dynasty. Tasan's allegoric poetry is inherited by Baek Seok[白石, 1912~1995] as regular juvenile literature. Baek Seok's juvenile stories are the results of expansion and transition for Tasan's allegoric poetry. Allegoric poetry was the shout of barber to prosecute about social irregularities and contradiction, and the sound of the bamboos to travel moaning of the people in the past. Now allegoric poetry create new emotion to make us to speculate ourselves with our surrounding. This changes are caused by special feature of allegoric poetry as a form to reflect our general lives.

• Proceedings of the IEEK Conference
• /
• 2002.07a
• /
• pp.94-97
• /
• 2002

A VLSI architecture of an Ogg Vorbis decoder is proposed : which is dedicated to portable audio appliances. Referring to the computational cost analysis of the decoding processes, the LSP (Line Spectrum Pair) process, which takes more than 50% of the total processing time, can be regarded as a bottleneck to achieve realtime processing by embedded Processors. Thus in our decoder a specific hardware architecture is devised for the LSP process so as to be integrated into a single chip together with an ARM7TDMI processor. In addition, in order to reduce the total hardware cost, instead of the floating point arithmetic, the fixed point arithmetic is adopted. The LSP module has been implemented with 9,740 gates by using a Virtual Silicon 0.l5$\mu\textrm{m}$ CMOS technology, which operates at 58.8MHz with the total CPU load reduced by 57%. It is also verified that the use of the fixed point arithmetic does not incur any significant sound distortion.

• Korean Journal of Materials Research
• /
• v.11 no.11
• /
• pp.986-990
• /
• 2001

An optimum route to fabricate the $A1_2O_3/Fe-Ni$ alloy nanocomposites with sound microstructure and enhanced mechanical properties as well as magnetism was investigated. To prepare homogeneous nanocomposite powders of Fe-Ni alloy and $Al_2O_3$, the solution-chemistry routes using $Al_2O_3 \; Ni(NO_3)_2{\cdot}6H_2O$ and $Fe(NO_3)_3{\cdot}9H_2O$ powders were applied. Microstructural observation of the powder mixture revealed that the Fe-Ni alloy particles of about 20 nm in size were homogeneously surrounded $A1_2O_3$, forming nanocomposite powder. The hot-pressed composite showed improved fracture toughness and magnetic response. These results suggest that the synergy materials with an improved mechanical properties and excellent functionality can be fabricated by controlled powder preparation and consolidation processing.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.5
• /
• pp.69-76
• /
• 1994

Acoustic cues of some consonantal phonology were tested in Korean words. All Korean consonants were recorded and acoustically analyzed in controlled phonological environments :ⅰ) word-initial, ⅱ) inter-vocalic, and ⅲ) word-final positions. The observed acoustic regulations are : ⅰ) The lengths of obstruents are longer word-initially than word-finally, ⅱ) The lengths of sonorants are longer word-finally than in word-initial or inter-vocalic positions, ⅲ) The formants of the lateral sound /l/ are higher word-finally than intervocalically. The phonological explanations of these acoustic regulations can be found in the rules of ⅰ) inter-vocalic voicing of plain stops, ⅱ) syllable-final unreleasing of obstruents, ⅲ) word-initial aspiration of stops, and ⅳ) liquid alternation between [r] and [l]. Numerical data of all these acoustic regulations are reported in order to facilitate their application toward improving naturalness for speech synthesis and accurateness for speech recognition.

• Proceedings of the IEEK Conference
• /
• 1999.06a
• /
• pp.559-562
• /
• 1999

In this paper, we developed on internet based assembly information display system for the blind. The system is consist of hardware and software. The hardware is consist of a voice synthesis device and a tactile display for character information, and the software is consist of internet web browser for the blind and braille program. The tactile-device system consists of a control unit, pin array, pin generator, serial port, and a power supply. The pin exerted by a electromagnetic method, solenoid. The internet web browser separates the character and image from internet web page, and character information in the web page is converted to braille and fed to sound system. Also the image in the web page can be printed developed tactile display. As the results of experiment, the blind could access the internet web site by using this system and understand various internet information.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.5
• /
• pp.1361-1369
• /
• 2000

For the implementation of 3D Sound Localization system, the binaural filtering by HRTFs is generally employed. But the HRTF filter is of high order and its coefficients for all directions have to be stored, which imposes a rather large memory requirement. To cope with this, research works have centered on obtaining low dimensional HRTF representations without significant loss of information and synthesizing the original HRTF efficiently, by means of feature extraction methods for multivariate dat including PCA. In these researches, conventional linear PCA was applied to the frequency domain HRTF data and using relatively small number of principal components the original HRTFs could be synthesized in approximation. In this paper we applied neural network based nonlinear PCA model (NLPCA) and the nonlinear PLS repression model (NLPLS) for this low dimensional HRTF modeling and analyze the results in comparison with the PCA. The NLPCA that performs projection of data onto the nonlinear surfaces showed the capability of more efficient HRTF feature extraction than linear PCA and the NLPLS regression model that incorporates the direction information in feature extraction yielded more stable results in synthesizing general HRTFs not included in the model training.