• MALSORI
• /
• no.37
• /
• pp.119-128
• /
• 1999

이 논문의 목적은 일본 학생들의 영어 청취 능력에 대한 확실한 평가를 내릴 수 있는 청취판별테스트를 개발하는 데에 있다. 이 테스트를 통하여 일본사람들이 범하는 청취 오류를 평가, 분석하고 일본어의 음성·음운체계가 이러한 오류에 미치는 영향을 평가한다. 테스트의 결과는 청취 및 발음훈련에 적용될 수 있으므로 일본인의 영어 능력을 향상시키는데 공헌할 수 있을 것이다.

• Proceedings of the KIEE Conference
• /
• 2006.10c
• /
• pp.419-421
• /
• 2006

The importance of automatic discrimination image data has evolved as a research topic over recent years. We have used forward neural network as a classifier using sound data features within image data, our initial tests have shown encouraging results that indicate the viability of our approach.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.2136-2139
• /
• 2003

본 논문은 FM 라디오 방송중의 오디오 신호를 블록단위로 음악 및 음성을 검출하는 알고리즘에 대한 것으로, 이를 기반으로 방송중의 노래(가요, 팝, 클래식‥‥)만을 자동으로 인식하여 녹음하는 알고리즘을 개발한다. 본 논문에서는 기존에 제안되었던 것[1-4]과 같이 단지 음악과 음성을 구분함과 동시에 음악구간의 논리적 조합으로 이루어진 노래를 자동으로 인식하여 녹음하는 것을 알고리즘의 최종 목표로 한다. 알고리즘의 접근 역시 기존의 음소단위의 모델링을 거치는 GMM 기반의 접근이 아니기 때문에 모델링에 대한 훈련과정이 필요 없고, 시간영역에서의 오디오신호가 가지고 있는 직관적인 특징을 분석함으로써 비교적 적은 연산으로 실시간 구현이 가능하다.

• Speech Sciences
• /
• v.14 no.3
• /
• pp.111-125
• /
• 2007

It is important to examine patients' subjective evaluation as well as objective measures and clinician's rating to assess voice disorders. This study aimed to evaluate validity and reliability of Korean-version of Voice Handicap Index (KVHI) and Voice-Related Quality of Life (KVQOL) with 113 adults with voice disorders and 111 normal adults. Content validity was verified by three experienced speech-language pathologists. Concurrent validity was revealed by examining the correlation among KVHI, KVQOL, and Voice Rating Scale as well as item discrimination coefficients. Total scores of KVHI and KVQOL of adults with voice disorders were significantly different from those of normal adults. Test-retest reliability and internal consistencies were significantly high in both KVHI and KVQOL. Correlations among scores of each subscale and total score were also significantly high in each tool. The study revealed that KVHI and KVQOL are suitable tools to be used in clinics and research areas in Korea, which can subjectively evaluate the effects of voice disorders on daily life as well as on quality of life.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.5
• /
• pp.81-88
• /
• 2010

In this paper, we proposed an automatic music/non-music signal discrimination system from broadcasting audio signal as a preliminary study of building a sound source monitoring system in real broadcasting environment. By reflecting human speech articulation characteristics, we used three simple time-domain features such as energy standard deviation, log energy standard deviation and log energy mean. Based on the experimental threshold values of each feature, we developed a rule-based algorithm to classify music portion of the input audio signal. For the verification of the proposed algorithm, actual FM broadcasting signal was recorded for 24 hours and used as source input audio signal. From the experimental results, the proposed system can effectively recognize music section with the accuracy of 96% and non-music section with that of 87%, where the performance is good enough to be used as a pre-process module for the a sound source monitoring system.

• The Journal of Korean Medicine
• /
• v.30 no.4
• /
• pp.169-178
• /
• 2009

Objectives: Sudden sensorineural hearing loss (SSHL) is considered an ENT emergency. Despite being a well-recognized condition, SSHL remains one of the most controversial issues in otology. Nowadays, more and more patients have an interest in Oriental medicine for treatment of SSHL. So, to ascertain the therapeutic effect of Oriental Medicine on SSHL, nineteen cases of SSHL patients who had taken Oriental medical therapy in Kyung Hee Oriental Medical Hospital were examined and analyzed. Methods: Nineteen patients who received over 10 times acupuncture therapy and a minimum 2 weeks of herbal medicine from Sep. 1, 2007 to Aug. 31, 2008 were examined and analyzed. The patients who were in the categories below were excluded: - within 7 days after onset - didn't fulfill 10 times acupuncture therapy - failed to recheck hearing outcome after treatment - less than 30dB at mean dB from 250Hz${\sim}$4000Hz. Results: The patients consisted of 12 men and 7 women with a mean age of 45.63 years (19${\sim}$76). Before treatment, 17 patients had tinnitus, 16 patients had pressure in the ear and 6 patients had dizziness, and mean dB of all patients was 66.89 dB. After treatment, 9 patients still had tinnitus, 4 patients felt pressure in the ear and 2 patents felt dizziness, and mean dB of all patients was 54.57dB. After treatment, 9 patients showed effectiveness in improving both hearing level and speech discrimination, 6 patients showed effectiveness only on speech discrimination and 4 patients showed no therapeutic effect. Conclusion: Oriental medical therapy had some therapeutic effects on SSHL even it was started 7 days after onset of the disease.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.1E
• /
• pp.30-38
• /
• 1995

In this paper, we evaluated the performance of vocoders for three-way conferencing using signal summation technique in digital mobile communication network. The signal summation technique yields natural mode of three-way conferencing, in shich the mixed voice signal from two speakers are transmitted to a third person, though there has been no useful speech coding technique for the mixed voice signal yet. We established Qualcomm code term prediction (RPE-LTP) vocoders to provide three-way conferencing using signal summation techinique. In addition, as the conventional speech quality measures are not applicable to the vocoders for mixed voice signals, we proposed two kinds of subjective quality measures. These are the sentence discrimination (SD) test and the modified degraded mean opinion score (MDMOS) test. The experimental results show that the output speech quality of the VSELP vocoder is superior to other two.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.5
• /
• pp.188-194
• /
• 2014

The feature extraction mechanism of the CASA(Computational Auditory Scene Analysis) system uses time continuity and frequency channel similarity to compose a correlogram of auditory elements. In segmentation, we compose a binary mask by using cross-correlation function, mask 1(speech) has the same periodicity and synchronization. However, when there is delay between autocorrelation signals with the same periodicity, it is determined as a speech, which is considered to be a drawback. In this paper, we proposed an algorithm to improve discrimination of channel similarity using Weighted Cross-correlation in segmentation. We conducted experiments to evaluate the speech segregation performance of the CASA system in background noise(siren, machine, white, car, crowd) environments by changing SNR 5dB and 0dB. In this paper, we compared the proposed algorithm to the conventional algorithm. The performance of the proposed algorithm has been improved as following: improvement of 2.75dB at SNR 5dB and 4.84dB at SNR 0dB for background noise environment.

• Journal of IKEEE
• /
• v.13 no.3
• /
• pp.62-71
• /
• 2009

In this study, we developed a wireless CROS(contralateral routing of signal) hearing aid for unilateral impaired people. CROS hearing aid takes sound from an ear with poorer hearing and transmit to another ear with better hearing. Generally, the self-voice delivered through the receiver of CROS hearing aid can be very loud. It is hard to perceive target speech because of loud self-voice. To compensate it, a self-voice suppression algorithm has been developed. we performed SDT(speech discrimination test) for evaluation of the self-voice suppression algorithm. One-syllable words was used as test speech and recorded with self-voice at a 1m distance. As the results, SDT score was improved about 11% when the self-voice suppression algorithm was processed. It is verified that the self-voice suppression algorithm helps speech perception at a time to communicate with others.

• Phonetics and Speech Sciences
• /
• v.14 no.4
• /
• pp.35-43
• /
• 2022

This paper studies how to improve the identification rate of laryngeal disability speech data by convolutional neural network (CNN) and machine learning ensemble learning methods. In general, the number of laryngeal dysfunction speech data is small, so even if identifiers are constructed by statistical methods, the phenomenon caused by overfitting depending on the training method can lead to a decrease the identification rate when exposed to external data. In this work, we try to combine results derived from CNN models and machine learning models with various accuracy in a multi-voting manner to ensure improved classification efficiency compared to the original trained models. The Pusan National University Hospital (PNUH) dataset was used to train and validate algorithms. The dataset contains normal voice and voice data of benign and malignant tumors. In the experiment, an attempt was made to distinguish between normal and benign tumors and malignant tumors. As a result of the experiment, the random forest method was found to be the best ensemble method and showed an identification rate of 85%.