• The Journal of the Acoustical Society of Korea
• /
• v.22 no.1E
• /
• pp.43-48
• /
• 2003

The effects of linear acoustic channels have been analyzed and compensated at mel-frequency feature domain. Unlike popular RASTA filtering our approach incorporates separate filters for each mel-frequency band, which results in better recognition performance for heavy-reverberated speeches.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.190-194
• /
• 2008

This paper presents a speaker identification technique which is one of the basic techniques of the ubiquitous robot companion. Though the conventional mel-frequency cepstral coefficients guarantee high performance of speaker identification in clean condition, the performance is degraded dramatically in noise condition. To overcome this problem, we employed the relative autocorrelation sequence mel-frequency cepstral coefficient which is one of the noise robust features. However, there are two problems in relative autocorrelation sequence mel-frequency cepstral coefficient: 1) the limited information problem. 2) the residual noise problem. In this paper, to deal with these drawbacks, we propose a multi-streaming method for the limited information problem and a hybrid method for the residual noise problem. To evaluate proposed methods, noisy speech is used in which air conditioner noise, classic music, and vacuum noise are artificially added. Through experiments, proposed methods provide better performance of speaker identification than the conventional methods.

• MALSORI
• /
• no.64
• /
• pp.89-103
• /
• 2007

In this paper, we introduce mel-cepstrum modulation energy (MCME) for a feature to discriminate speech and music data. MCME is a mel-cepstrum domain extension of modulation energy (ME). MCME is extracted on the time trajectory of Mel-frequency cepstral coefficients, while ME is based on the spectrum. As cepstral coefficients are mutually uncorrelated, we expect the MCME to perform better than the ME. To find out the best modulation frequency for MCME, we perform experiments with 4 Hz to 20 Hz modulation frequency. To show effectiveness of the proposed feature, MCME, we compare the discrimination accuracy with the results obtained from the ME and the cepstral flux.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.12
• /
• pp.1137-1142
• /
• 2010

This paper studied the speech parameters less affected by the human emotion for the development of the robust speech recognition system. For this purpose, the effect of emotion on the speech recognition system and robust speech parameters of speech recognition system were studied using speech database containing various emotions. In this study, mel-cepstral coefficient, delta-cepstral coefficient, RASTA mel-cepstral coefficient and frequency warped mel-cepstral coefficient were used as feature parameters. And CMS (Cepstral Mean Subtraction) method were used as a signal bias removal technique. Experimental results showed that the HMM based speaker independent word recognizer using vocal tract length normalized mel-cepstral coefficient, its derivatives and CMS as a signal bias removal showed the best performance of 0.78% word error rate. This corresponds to about a 50% word error reduction as compare to the performance of baseline system using mel-cepstral coefficient, its derivatives and CMS.

• Applied Chemistry for Engineering
• /
• v.32 no.3
• /
• pp.299-304
• /
• 2021

This study focuses on the preparation of acetaminophen (AP) imprinted functional biomaterials for a transdermal drug delivery using mung bean starch (MBS), polyvinyl alcohol (PVA), sodium benzoate (S) as a crosslinking agent, glycerol (GL) as a plasticizer, and melanin (MEL) as a photothermal agent. The prepared AP imprinted biomaterials were characterized using FE-SEM and their physical properties were evaluated. The photothermal effect and AP release property for functional biomaterials were examined with the irradiation of near infrared (NIR) laser (1.5 W/cm²). When the NIR laser was irradiated on functional biomaterials with/without the addition of MEL, the temperature of MEL added biomaterial increased from 25 ℃ to 41 ℃, whereas the biomaterial without MEL increased from 25 ℃ to 28 ℃. Results indicate that there is the photothermal effect of prepared biomaterial with the addition of MEL. Based on the results, AP release properties were evaluated using standard buffer solutions and artificial skin. It was found that AP release rates of MEL added AP loaded biomaterials were 1.2 times faster than those of MEL non-added AP loaded biomaterials when irradiating with NIR laser. We envision that the developed functional biomaterials can be utilized for an acute pain-killing treatment.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.3
• /
• pp.143-149
• /
• 2020

This paper presents the effect of the feature extraction methods used in the audio preprocessing on the classification performance of the Convolutional Neural Networks (CNN). We extract mel spectrogram, log mel spectrogram, Mel Frequency Cepstral Coefficient (MFCC), and delta MFCC from the UrbanSound8K dataset, which is widely used in environmental sound classification studies. Then we scale the data to 3 distributions. Using the data, we test four CNNs, VGG16, and MobileNetV2 networks for performance assessment according to the audio features and scaling. The highest recognition rate is achieved when using the unscaled log mel spectrum as the audio features. Although this result is not appropriate for all audio recognition problems but is useful for classifying the environmental sounds included in the Urbansound8K.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1157-1157
• /
• 2001

Yeast, Kurtzurnanomyces sp. I-11, produces biosurfactant, mannosyl erythritol lipid (MEL), from soya oil. The properties of biosurfactant MEL include low-toxicity and high biodegradability. MEL provides new possibilities for a wide range of industrial applications, especially food, cosmetic, pharmaceutical fields and chemicals for biotechnology. In the fermentation process, techniques of measuring and controlling substrates and products are important to obtain high productivity with optimum concentrations of substrate and product in the culture broth. The measurement system for the concentrations of soya oil and MEL in the fermentation process was developed using near-infrared spectroscopy (NIRS). Soya oil and MEL in the culture broth were extracted with ethyl acetate and NIR spectra was carried out between the second derivative NIR spectral data at 1312 and 2040 nm and MEL concentrations obtained using a thin-layer chromatography with a flame-ionization detector (TLC/FID) method. A calibration equation for soya oil was results of the validation of the calibration equation, good agreement was observed between the results of the TLD/FID method and those of the NIRS method for both constituents. NIR method was applied to the measurement of the concentrations of MEL and soya oil in the practical fermentation and good results were obtained. The study indicates that NIRS is a useful method for measurement of the substrate and product in the glycolipid fermentation.

• YAKHAK HOEJI
• /
• v.48 no.2
• /
• pp.147-152
• /
• 2004

A wide variety of cancer chemotherapeutic agents have been shown to induce programmed cell death (PCD, APOPTOSIS) in various tumor cell lines in vitro. cis-Malonato [(4R,5R)-4,5-bis(aminomethyl)-2-isoprpopyl-1,3-dioxolane] platinum(II) (heptaplatin), which is a new drug approved by KFDA in 1999, in a novel platinum-based antitumor agent with clinical potential against stomach cancer and the 3rd generation of the cisplatin. This study was performed to know how heptaplatin and cisplatin and sunpla (mixture of heptaplatin and mannitol) affect on SK-MEL-28 cell line, and how they induce the apoptosis. At EM analysis, the morphology of the cell was changed by treatment of the cisplatin, heptaplatin and sunpla. Apoptotic body formed around plasma membrane, and chromatin condensation represented in nucleus. This phenomenon is one of the characteristic of the apoptosis. The DNA of SK-MEL-28 cell line truncated by cisplatin and sunpla treatment was identified on 2% agarose gel electrophoresis. TUNEL assay was performed to know whether SK-MEL-28 cell die as apoptosis or necrosis by cisplatin, heptaplatin and sunpla. At this result, fluorescence intensity increased according to increase of time and concentration. Therefore, it was identified that cislatin, heptaplatin and sunpla induced apoptosis. Fas expressed on SK-MEL-28 cell membrane by cisplatin, heptaplatin and sunpla was identified by using flow cytometer and the expression of bcl-2(anti-apoptotic gene) decreased according to increase of concentration of the cisplatin, heptaplatin and sunpla. Cisplatin, heptaplatin and sunpla induced apoptosis against SK-MEL-28 cell line, and the apoptotic mechanism was identified as Fas-mediated apoptosis and decreased bcl-2 expression.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.1
• /
• pp.39-46
• /
• 2019

In this paper, combining features is proposed as a way to enhance the classification accuracy of sounds under noisy environments using the CNN (Convolutional Neural Network) structure. A robust log Mel-filter bank using Wiener filter and PNCCs (Power Normalized Cepstral Coefficients) are extracted to form a 2-dimensional feature that is used as input to the CNN structure. An ebird database is used to classify 43 types of bird species in their natural environment. To evaluate the performance of the combined features under noisy environments, the database is augmented with 3 types of noise under 4 different SNRs (Signal to Noise Ratios) (20 dB, 10 dB, 5 dB, 0 dB). The combined feature is compared to the log Mel-filter bank with and without incorporating the Wiener filter and the PNCCs. The combined feature is shown to outperform the other mentioned features under clean environments with a 1.34 % increase in overall average accuracy. Additionally, the accuracy under noisy environments at the 4 SNR levels is increased by 1.06 % and 0.65 % for shop and schoolyard noise backgrounds, respectively.

• Journal of Life Science
• /
• v.20 no.5
• /
• pp.655-661
• /
• 2010

To elucidate the mechanism underlying the suppressive regulation of hST8Sia I expression in retinoic acid (RA)-induced SK-MEL-2 cells, we characterized the promoter region of the hST8Sia I gene. Functional analysis of the 5‘-flanking region of the hST8Sia I gene by the transient expression method showed that the -1146 to -646 region, which contains putative binding sites for transcription factors c-Ets-1, CREB, AP-1 and NF-kB, functions as the RA-repressive promoter in SK-MEL-2 cells. Site-directed mutagenesis and ChIP analyses indicated that the NF-kB binding site at -731 to -722 is crucial for the RA-induced repression of hST8Sia I in SK-MEL-2 cells. In addition, the transcriptional activity of hST8Sia I suppressed by RA in SK-MEL-2 cells was strongly inhibited by extracellular signal-regulated protein kinase (ERK) inhibitor U0126 and protein kinase C (PKC) inhibitor GO6976, as determined by RT-PCR and luciferase assay of hST8Sia I promoter containing the -1146 to -646 regions. These results suggest that RA markedly modulates transcriptional regulation of hST8Sia I gene expression through the PKC/ERK signal pathway in SK-MEL-2 cells.