• 한국음향학회지
• /
• 제22권5호
• /
• pp.388-402
• /
• 2003

In this paper, we carried out the word, 4 continuous digits. continuous, and task-independent word recognition experiments to verify the effectiveness of the re-defined phoneme-likely units (PLUs) for the phonetic decision tree based HM-Net (Hidden Markov Network) context-dependent (CD) acoustic modeling in Korean appropriately. In case of the 48 PLUs, the phonemes /ㅂ/, /ㄷ/, /ㄱ/ are separated by initial sound, medial vowel, final consonant, and the consonants /ㄹ/, /ㅈ/, /ㅎ/ are also separated by initial sound, final consonant according to the position of syllable, word, and sentence, respectively. In this paper. therefore, we re-define the 39 PLUs by unifying the one phoneme in the separated initial sound, medial vowel, and final consonant of the 48 PLUs to construct the CD acoustic models effectively. Through the experimental results using the re-defined 39 PLUs, in word recognition experiments with the context-independent (CI) acoustic models, the 48 PLUs has an average of 7.06%, higher recognition accuracy than the 39 PLUs used. But in the speaker-independent word recognition experiments with the CD acoustic models, the 39 PLUs has an average of 0.61% better recognition accuracy than the 48 PLUs used. In the 4 continuous digits recognition experiments with the liaison phenomena. the 39 PLUs has also an average of 6.55% higher recognition accuracy. And then, in continuous speech recognition experiments, the 39 PLUs has an average of 15.08% better recognition accuracy than the 48 PLUs used too. Finally, though the 48, 39 PLUs have the lower recognition accuracy, the 39 PLUs has an average of 1.17% higher recognition characteristic than the 48 PLUs used in the task-independent word recognition experiments according to the unknown contextual factor. Through the above experiments, we verified the effectiveness of the re-defined 39 PLUs compared to the 48PLUs to construct the CD acoustic models in this paper.

• Journal of Mechanical Science and Technology
• /
• 제20권12호
• /
• pp.2079-2086
• /
• 2006

A fundamental study for developing a fault diagnosis system of a pump is performed by using neural network. Acoustic signals were obtained and converted to frequency domain for normal products and artificially deformed products. The neural network model used in this study was 3-layer type composed of input, hidden, and output layer. The normalized amplitudes at the multiples of real driving frequency were chosen as units of input layer. And the codes of pump malfunctions were selected as units of output layer. Various sets of teach signals made from original data by eliminating some random cases were used in the training. The average errors were approximately proportional to the number of untaught data. Neural network trained by acoustic signals can detect malfunction or diagnose fault of a given machine from the results.

• 한국건축시공학회지
• /
• 제16권1호
• /
• pp.87-96
• /
• 2016

More than half of Korea's urban population currently lives in multi-unit dwellings. This particular residence structure inevitably positions the living spaces of residents in close proximity to one another. Because of this proximity, the sounds of a particular household will transfer to other residents' units creating conflict between tenants. Efforts to alleviate this problem have been made in both the public and private sector. A prominent method of noise analysis between living units has been to make an actual sound measurement and obtain qualitative measurements from questionnaires. Although this method has been most widely used, such analysis requires a large amount of human effort and time. In addition, the questionnaire method fails to provide the objective information needed to accurately assess the noise situation. To overcome the inadequacies of this previous method, this study seeks to evaluate the applicability of an acoustic simulation method. Three types of living noises were analyzed to assess the noise levels passed between multi-unit dwellings. The acoustic simulation method was found to be more economical, efficient, and adaptable in information processing. The results of this study can be further applied to design and control living-noise through procedures such as acoustic absorption or space redesign.

• The Journal of the Acoustical Society of Korea
• /
• 제20권2E호
• /
• pp.3-8
• /
• 2001

In this paper, we will examine a Korean large vocabulary continuous speech recognition (LVCSR) system for broadcast news speech. The combined vowel and implosive unit is included in a phone set together with other short phone units in order to obtain a longer unit acoustic model. The effect of this unit is compared with conventional phone units. The dictionary units for language processing are automatically extracted from eojeols appearing in transcriptions. Triphone models are used for acoustic modeling and a trigram model is used for language modeling. Among three major speaker groups in news broadcasts-anchors, journalists and people (those other than anchors or journalists, who are being interviewed), the speech of anchors and journalists, which has a lot of noise, was used for testing and recognition.

• 대한음성학회지:말소리
• /
• 제66호
• /
• pp.117-130
• /
• 2008

This paper presents a method for automatically detecting Korean prosodic boundaries using both acoustic and grammatical information for the performance improvement of speech information processing systems. While most of previous works are solely based on grammatical information, our method utilizes not only grammatical information constructed by a Maximum-Entropy-based grammar model using 10 grammatical features, but also acoustical information constructed by a GMM-based acoustic model using 14 acoustic features. Given that Korean prosodic structure has two intonationally defined prosodic units, intonation phrase (IP) and accentual phrase (AP), experimental results show that the detection rate of AP boundaries is 82.6%, which is higher than the labeler agreement rate in hand transcribing, and that the detection rate of IP boundaries is 88.7%, which is slightly lower than the labeler agreement rate.

• 한국소음진동공학회논문집
• /
• 제22권7호
• /
• pp.676-684
• /
• 2012

Numerical analysis was performed to investigate the heavy-weight impact noise of apartment houses. The FEM is practical method for prediction of low-frequency indoor noise. The results of numerical analysis, the shape of the acoustic modes in room-2 are similar to that of acoustic pressure field at the fundamental frequency of acoustic modes. And the acoustic pressure was amplified at the natural frequency of the acoustic modes and structural modes. The numerical analysis result of sound pressure level at 63 Hz and 125 Hz octave-band center frequency are similar to the test results, but at 250 Hz and 500 Hz have some errors. Considering most of bang-machine force spectrum exists below 100 Hz, the noise at 250 Hz and 500 Hz are not important for heavy-weight impact noise. Thus, the FEM numerical analysis method for heavy-weight impact noise can apply to estimate heavy-weight impact noise for various building systems.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2003년도 추계학술대회
• /
• pp.137-142
• /
• 2003

A fundamental study for developing a system of fault diagnosis of a pump is performed by using neural network. The acoustic signals were obtained and converted to frequency domain for normal products and artificially deformed products. The signals were obtained in various driving frequencies in order to obtain many types of data from a limited number of pumps. The acoustic data in frequency domain were managed to multiples of real driving frequency with the aim of easy comparison. The neural network model used in this study was 3-layer type composed of input, hidden, and output layer. The normalized amplitudes at the multiples of real driving frequency were chosen as units of input layer, Various sets of teach signals made from original data by eliminating some random cases were used in the training. The average errors were approximately proportional to the number of untaught data. The results showed neural network trained by acoustic signals can be used as a simple method far a detection of machine malfunction or fault diagnosis.

• 한국공작기계학회논문집
• /
• 제13권5호
• /
• pp.81-86
• /
• 2004

A fundamental study for developing a system of fault diagnosis of a pump is performed by using neural network. The acoustic signals were obtained and converted to frequency domain for normal products and artificially deformed products. The signals were obtained in various driving frequencies in order to obtain many types of data from a limited number of pumps. The acoustic data in frequency domain were managed to multiples of real driving frequency with the aim of easy comparison. The neural network model used in this study was 3-layer type composed of input, hidden, and output layer. The normalized amplitudes at the multiples of real driving frequency were chosen as units of input layer. Various sets of teach signals made from original data by eliminating some random cases were used in the training. The average errors were approximately proportional to the number of untaught data. The results showed neural network trained by acoustic signals can be used as a simple method for a detection of machine malfuction or fault diagnosis.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2001년도 추계학술대회논문집 II
• /
• pp.1306-1312
• /
• 2001

In this study. the acoustic power balancing method to analysis HVAC system noise is presented. The method can consider not only forward but also backward propagations of noise generated by the operation of air supply units and aerodynamical disturbance at duct elements. This can be done by estimating sound transmission and reflection properties of duct elements. and balancing acoustic powers of total HVAC system. To verify the accuracy of the presented method. numerical analysis for a HVAC system is carried out and the results are compared with those obtained by a traditional empirical method. suggested by National Environmental Balancing Bureau.

• 말소리와 음성과학
• /
• 제1권3호
• /
• pp.117-122
• /
• 2009

A variety of cross-linguistic studies has documented that the acoustic properties of speech addressed to young children include exaggeration of pitch contours and acoustically salient features of phonetic units. It has been suggested that phonetic modifications of child-directed speech facilitate young children's learning of speech sounds by providing detailed phonetic information about the target word. While there are several studies reporting vowel modifications in speech to infants (i.e., hyper-articulated vowels), there has been little research about consonant modifications in speech to young children (except for VOT). The present study examines acoustic properties of Korean stops in Korean mothers' speech to their children (seven children aged 27 to 38 months). Korean tense, lax, and aspirated stops are all voiceless in word-initial position, and are perceptually differentiated by several acoustic parameters including VOT, $f_0$ of the following vowel, and the amplitude difference of the first and second harmonics at the voice onset of the following vowel. This study compares values of these parameters in Korean child-directed speech to those in adult-directed speech from same speakers. Conclusions focus on the acoustic properties of Korean stops in child-directed speech and how they are modified to help Korean young children learn the three-way phonetic contrast.