• Journal of the Korean Society of Laryngology, Phoniatrics and Logopedics
• /
• v.32 no.2
• /
• pp.87-93
• /
• 2021

There are a number of voice analysis programs around the world. Domestic voice analysis is performed by relying heavily on specific commercial program. We intend to develop coding for voice analysis using Praat and apply it to clinical practice. This study consisted of Experiment 1 and Experiment 2. Experiment 1 was the development of automated voice analysis coding based on Praat. The coding was largely divided into a recording, an analysis, and a storage section. Experiment 2 was applied to the voice analysis of 2 male patients pre- and post-operation with this coding. The analysis parameters of this coding provided 26 parameters for vowel /a/, nine parameters for sentence analysis, and a total of 4 parameters for voice range profile analysis. In two male patients, the pitch and the intensity increased, the voice quality improved, and the sentence length decreased after surgery. The coding was well made, so the output was good in real time. The code is automated as much as possible to block manual errors and increases convenience and efficiency by generating the result sheet in real time.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.1
• /
• pp.341-348
• /
• 2022

Formulaic sequences are continuous or discontinuous series of words that are seemingly treated like single units. Formulaic sequences play a key role in language development, and formulaic sequences acquisition determines the success or failure of language development. This study proposes a reading aloud activity as a way for international students to learn formulaic sequences. A class focused on reading aloud was conducted with 41 international students taking a general English course at a university in Seoul. For 15 weeks, video lectures and real-time Zoom classes were conducted in parallel. The animated film Frozen was used as course material. In the video lectures, the teacher interpreted the movie script in easy Korean and read aloud formulaic sequences. Students were tasked with reading the sentences with formulaic sequences aloud, recording themselves reading aloud, and submitting their recordings. During real-time class meetings, students performed the activity of reading aloud the formulaic sequences they had studied in the video lectures. There was a significant increase in the interpretation and sentence writing of formulaic sequences in participants' post-evaluation compared to the pre-evaluation. Through the study's survey, students exhibited positive views in the affective domains.

• Korean Journal of Applied Biomechanics
• /
• v.17 no.1
• /
• pp.81-90
• /
• 2007

The effort identifying positioning information of the moving object in real time has been a issue not only in sport biomechanics but also other academic areas. In order to solve this issue, this study tried to track the movement of a pitched ball that might provide an easier prediction because of a clear focus and simple movement of the object. Machine learning has been leading the research of extracting information from continuous images such as object tracking. Though the rule-based methods in artificial intelligence prevailed for decades, it has evolved into the methods of statistical approach that finds the maximum a posterior location in the image. The development of machine learning, accompanied by the development of recording technology and computational power of computer, made it possible to extract the trajectory of pitched baseball from recorded images. We present a method of baseball tracking, based on object tracking methods in machine learning. We introduce three state-of-the-art researches regarding the object tracking and show how we can combine these researches to yield a novel engine that finds trajectory from continuous pitching images. The first research is about mean shift method which finds the mode of a supposed continuous distribution from a set of data. The second research is about the research that explains how we can find the mode and object region effectively when we are given the previous image's location of object and the region. The third is about the research of representing data into features that we can deal with. From those features, we can establish a distribution to generate a set of data for mean shift. In this paper, we combine three works to track baseball's location in the continuous image frames. From the information of locations from two sets of images, we can reconstruct the real 3-D trajectory of pitched ball. We show how this works in real pitching images.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.2
• /
• pp.102-111
• /
• 2023

Automatic Music Transcription (AMT) is a task that detects and recognizes musical note events from a given audio recording. In this paper, we focus on reducing the latency of real-time AMT systems on piano music. Although neural AMT models have been adapted for real-time piano transcription, they suffer from high latency, which hinders their usefulness in interactive scenarios. To tackle this issue, we explore several techniques for reducing the intrinsic latency of a neural network for piano transcription, including reducing window and hop sizes of Fast Fourier Transformation (FFT), modifying convolutional layer's kernel size, and shifting the label in the time-axis to train the model to predict onset earlier. Our experiments demonstrate that combining these approaches can lower latency while maintaining high transcription accuracy. Specifically, our modified model achieved note F1 scores of 92.67 % and 90.51 % with latencies of 96 ms and 64 ms, respectively, compared to the baseline model's note F1 score of 93.43 % with a latency of 160 ms. This methodology has potential for training AMT models for various interactive scenarios, including providing real-time feedback for piano education.

• Journal of Sensor Science and Technology
• /
• v.17 no.2
• /
• pp.87-94
• /
• 2008

Blood pressure (BP), one of the most important vital signs, is used to identify an emergency state and reflects the blood flow characteristics of the cardiovascular system. The conventional noninvasive method of measuring BP is inconvenient because patients must wear a cuff on their arm and the measurement process takes time. This paper proposes an algorithm for estimating the BP using the pulse transit time (PTT) of the photoplethysmography (PPG) and pressure pulse from finger at the same time as a more convenient way to measure the BP. After recording the electrocardiogram (ECG), measuring the pressure pulse, and performing PPG, we calculated the PTT from the acquired signals. Then, we used a multiple regression analysis to measure the systolic and diastolic BP indirectly. Comparing the BP measured indirectly using the proposed algorithm and the real BP measured with a sphygmomanometer, the systolic pressure had a mean error of ${\pm}3.240$ mmHg and a standard deviation of 2.530 mmHg, while the diastolic pressure had a satisfactory result, i.e., a mean error of ${\pm}1.807$ mmHg and a standard deviation of 1.396 mmHg. These results are more superior than existing method estimating blood pressure using the one PTT and satisfy the ANSI/AAMI regulations for certifying a sphygmomanometer i.e., the measurement error should be within a mean error of ${\pm}5$ mmHg and a standard deviation of 8 mmHg. These results suggest the possibility of applying our method to a portable, long-term BP monitoring system.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.304-308
• /
• 2008

In this paper, we present measurement method of the amount of used piped water and the communications environment for automatic piped water meter reading. The existing piped water meter reading which is carried out by the human in person by monitoring the amount of use and recording make the problem of an error of the figure and being short of staffed to carry out piped water meter reading. To solve the existing problem in this paper, the amount of used piped water is converted into data type by Photo Interrupt Sensor, and IEEE 802.15.4 Protocol is used to confirm the measurement in remote place.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.10a
• /
• pp.846-849
• /
• 2008

In recent years there has been growing interest in wireless sensor networks (WSNs) for a variety of indoor applications. In this paper, we present the RSSI-based localization in indoor environments. In order to evaluate the relationship between distance and RSSI, the log-normal path loss shadowing model is used. By tagging users with a sensor node and deploying a number of nodes at fixed position in the building, the RSSI can be used to determine the position of tagged user. This system operates by recording and processing signal strength information at the base stations. It combines Euclidean distance technique with signal strength matrix obtained during real-time measurement to determine the location of user. The experimental results presented the ability of this system to estimate user's location with a accuracy.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.607-609
• /
• 2005

본 논문에서는 실시간으로 다채널의 카메라에서 입력되는 영상을 동시에 압축 및 복원할 수 있는 PC 기반의 DVR (Digital Video Recording) 시스템을 위한 MPEG-4 방식의 코덱을 구현하고자 한다. 현재까지의 일반적인 압축 방식은 화상회의 용도의 H.263, VCD 화질의 MPEG-1, DVD급 화질인 MPEG-2가 널리 적용되고 있다. 하지만 이러한 방법들은 저장 데이터의 양이 커서 효율적인 저장이 어렵다. 따라서 본 논문에서는 이러한 문제점을 해결하면서 실시간적으로 다채널 영상 데이터 저장이 가능한 MPEG-4 압축 방식을 적용한 코덱을 제작한다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.10a
• /
• pp.157-160
• /
• 2011

In this paper, We proposed the method of recognition and analysis of human activites using Motion vector in real-time surveillance system. We employs subtraction image techniques to detect blob(human) in the foreground. When MPEG-4 video recording EPZS(Enhanced Predicted Zonal Search) is detected the values of motion vectors were used. In this paper, the activities of human recognize and classified such as meta-classes like this {Active, Inactive}, {Moving, Non-moving}, {Walking, Running}. Each step was separated using a step-by-step threshold values. We created approximately 150 conditions for the simulation. As a result, We showed a high success rate about 86~98% to distinguish each steps in simulation image.

• Journal of Biomedical Engineering Research
• /
• v.21 no.5
• /
• pp.469-475
• /
• 2000

신경신호의 계측은 신경계의 연구에 필수적인 도구로 최근 반도체미세전극기술 등 수십, 수백개의 채널로부터 신경신호를 기록할 수 있는 방법들이 발달함에 따라 많은 수의 뉴런으로부터 신경 신호를 측정하여 컴퓨터로 그 신호를 처리할 수 있는 시스템의 필요성은 더욱 커지고 있다. 본 연구에서는 최대 16채널의 신경신호를 실시간에 측정하여 기록하고, 저장된 신호로부터 활동전위를 검출하며, 단일 뉴런들로부터의 신호를 분류하여 spike train의 형태로 저장한 뒤 여러 뉴런들간의 상관관계를 분석하기 위한 spike train 해석이 가능한 시스템을 개발하였다. 이 시스템은 보통사양의 PC이외에는 단지 신호획득보드만을 포함하여 다채널미세전극으로부터 뉴런의 신호를 측정, 증폭하여 호스트PC로 전송하고 저장하며 이로부터 활동전위를 검출하여 단일뉴런으로부터의 spike train으로 분류할 수 있다. 또한 저장된 spike train들로부터 신경회로망을 이루는 여러뉴런 들간의 관계를 분석하여 기능들이 시스템에 포함되어있다. 개발된 시스템을 사용하여 개구리 감각 신경의 신호를 실시간에 동시기록하여 활동전위을 검출하고 특징추출방법과 principal component analysis를 이용하여 분류한 뒤 spike train 해석을 수행하였다.