• Phonetics and Speech Sciences
• /
• v.13 no.2
• /
• pp.57-66
• /
• 2021

This study focuses on the issue of automatic severity classification of dysarthric speakers based on speech intelligibility. Speech intelligibility is a complex measure that is affected by the features of multiple speech dimensions. However, most previous studies are restricted to using features from a single speech dimension. To effectively capture the characteristics of the speech disorder, we extracted features of multiple speech dimensions: voice quality, prosody, and pronunciation. Voice quality consists of jitter, shimmer, Harmonic to Noise Ratio (HNR), number of voice breaks, and degree of voice breaks. Prosody includes speech rate (total duration, speech duration, speaking rate, articulation rate), pitch (F0 mean/std/min/max/med/25quartile/75 quartile), and rhythm (%V, deltas, Varcos, rPVIs, nPVIs). Pronunciation contains Percentage of Correct Phonemes (Percentage of Correct Consonants/Vowels/Total phonemes) and degree of vowel distortion (Vowel Space Area, Formant Centralized Ratio, Vowel Articulatory Index, F2-Ratio). Experiments were conducted using various feature combinations. The experimental results indicate that using features from all three speech dimensions gives the best result, with a 80.15 F1-score, compared to using features from just one or two speech dimensions. The result implies voice quality, prosody, and pronunciation features should all be considered in automatic severity classification of dysarthria.

• Smart Media Journal
• /
• v.10 no.3
• /
• pp.31-38
• /
• 2021

It is hard to predict when and where a fall accident will happen. Also, if rapid follow-up measures on it are not performed, a fall accident leads to a threat of life, so studies that can automatically detect a fall accident have become necessary. Among automatic fall-accident detection techniques, a fall detection scheme using an IMU (inertial measurement unit) sensor attached to a wrist is difficult to detect a fall accident due to its movement, but it is recognized as a technique that is easy to wear and has excellent accessibility. To overcome the difficulty in obtaining fall data, this study proposes an algorithm that efficiently learns less data through machine learning such as KNN (k-nearest neighbors) and SVM (support vector machine). In addition, to improve the performance of these mathematical classifiers, this study utilized feature data aquired in the frequency space. The proposed algorithm analyzed the effect by diversifying the parameters of the model and the parameters of the frequency feature extractor through experiments using standard datasets. The proposed algorithm could adequately cope with a realistic problem that fall data are difficult to obtain. Because it is lighter than other classifiers, this algorithm was also easy to implement in small embedded systems where SIMD (single instruction multiple data) processing devices were difficult to mount.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2023.05a
• /
• pp.100-103
• /
• 2023

Recently, development of maritime autonomoust surface ships and eco-friendly ships, production and evaluation research considering various marine environments is needed in the field of optimal routes as the demand for accurate and detailed real-time marine environment prediction information expands. An algorithm that can calculate the optimal route while reducing the risk of the marine environment and uncertainty in energy consumption in smart ships was developed in 2 stages. In the first stage, a profile was created by combining marine environmental information with ship location and status information within the Automatic Ship Identification System(AIS). In the second stage, a model was developed that could define the marine environment energy map using the configured profile results, A regression equation was generated by applying Random Forest among machine learning techniques to reflect about 600,000 data. The Random Forest coefficient of determination (R²) was 0.89, showing very high reliability. The Dijikstra shortest path algorithm was applied to the marine environment prediction at June 1 to 3, 2021, and to calculate the optimal safety route and express it on the map. The route calculated by the random forest regression model was streamlined, and the route was derived considering the state of the marine environment prediction information. The concept of route calculation based on real-time marine environment prediction information in this study is expected to be able to calculate a realistic and safe route that reflects the movement tendency of ships, and to be expanded to a range of economic, safety, and eco-friendliness evaluation models in the future.

• Journal of Convergence for Information Technology
• /
• v.11 no.3
• /
• pp.1-6
• /
• 2021

Most of the match-3 puzzle games supports automatic play using the MCTS algorithm. However, implementing reinforcement learning agents is not an easy job because it requires both the knowledge of machine learning and the way of complex interactions within the development environment. This study proposes a method in which we can easily design reinforcement learning agents and implement game play agents by applying PPO(Proximal Policy Optimization) algorithms. And we could identify the performance was increased about 44% than the conventional method. The tools we used are the Unity 3D game engine and Unity ML SDK. The experimental result shows that agents became to learn game rules and make better strategic decisions as experiments go on. On average, the puzzle gameplay agents implemented in this study played puzzle games better than normal people. It is expected that the designed agent could be used to speed up the game level design process.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.281-283
• /
• 2021

The representative companies mentioned in the recommendation service in the domestic OTT(Over-the-top media service) market are YouTube and Netflix. YouTube, through various methods, started personalized recommendations in earnest by introducing an algorithm to machine learning that records and uses users' viewing time from 2016. Netflix categorizes users by collecting information such as the user's selected video, viewing time zone, and video viewing device, and groups people with similar viewing patterns into the same group. It records and uses the information collected from the user and the tag information attached to the video. In this paper, we propose a method to improve video media recommendation by automatically generating metadata of video media that was written by hand.

• Phonetics and Speech Sciences
• /
• v.15 no.3
• /
• pp.75-82
• /
• 2023

Automatic speech recognition (ASR) has been revolutionized with deep learning-based approaches, among which self-supervised learning methods have proven to be particularly effective. In this study, we aim to enhance the performance of OpenAI's Whisper model, a multilingual ASR system on the Korean language. Whisper was pretrained on a large corpus (around 680,000 hours) of web speech data and has demonstrated strong recognition performance for major languages. However, it faces challenges in recognizing languages such as Korean, which is not major language while training. We address this issue by fine-tuning the Whisper model with an additional dataset comprising about 1,000 hours of Korean speech. We also compare its performance against a Transformer model that was trained from scratch using the same dataset. Our results indicate that fine-tuning the Whisper model significantly improved its Korean speech recognition capabilities in terms of character error rate (CER). Specifically, the performance improved with increasing model size. However, the Whisper model's performance on English deteriorated post fine-tuning, emphasizing the need for further research to develop robust multilingual models. Our study demonstrates the potential of utilizing a fine-tuned Whisper model for Korean ASR applications. Future work will focus on multilingual recognition and optimization for real-time inference.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.8
• /
• pp.873-877
• /
• 2010

CAPTCHA is a security tool that prevents the automatic sign-up by a spam or a robot. This CAPTCHA usually depends on the smart readability of humans. However, the common and plain CAPTCHA with text-based system is not difficult to be solved by intelligent web-bot and machine learning tools. In this paper, we propose a new sub-image based CAPTCHA system totally different from the text based system. Our system offers a set of cropped sub-image from a whole digital picture and asks user to identify the correct orientation. Though there are some nice machine learning tools for this job, but they are useless for a cropped sub-images, which was clearly revealed by our experiment. Experiment showed that our sub-image based CAPTCHA is easy to human solver, but very hard to all kinds of machine learning or AI tools. Also our CAPTCHA is easy to be generated automatical without any human intervention.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.2
• /
• pp.125-130
• /
• 2021

Implementation of livestock smart-farm can be done more effectively with IoT technology developing. An build of useful stock management system can be possibile if push messages of these judgement are notified on smart-phone after cattle's illness and estrus are judged using IoT technology. These judgement method of cattle's illness and estrus can be done with gathering living stock data using temperature sensor and 3 axis acceleration sensor and sending these data using IoT and internet network into server, and studying AI machine learning using these data. In this paper, to build this cattle management system based on IoT, effective system of the whole architecture is showed. Also an effective analysis and design method to develop this system software will be presented by showing user requirement analysis using object-oriented method, flowchart and screen design.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.400-400
• /
• 2023

급격한 기후변화로 인해 일사량, 지표면 온도 및 이산화탄소 농도가 꾸준히 상승함에 따라 수문 순환의 불균형을 초래함과 하천 및 호소 내 수질 또한 악화되고 있는 추세이다. 특히, 국내의 경우, 기후변화 및 인위적 요인에 의해 하천 및 호소에서의 수위 감소 및 수온 증가로 인해 부영양화가 증가되고 있고, 이로 인한 유해 녹조의 발생빈도를 높이는 결과를 초래한다. 현재 국내에서는 유인 수질 관측 및 자동 수질관측 시스템을 통해 주요 수질인자를 모니터링 하고 있으나 시·공간적인 변동성을 파악하는데 제한점이 있다. 이러한 한계점을 극복하기 위해 국·내외에서 광학위성을 이용한 수질인자 추정 알고리즘 개발과 관련된 연구들이 진행되고 있다. 이에 따라, 본 연구에서는 NASA에서 제공하는 Landsat-8 위성과 ESA에서 제공하는 Sentinel-2자료가 동화된 Harmonized Landsat Sentinel-2 위성자료를 활용한 클로로필-a (Chl-a)를 추정하고자 한다. 이를 위해, 본 연구에서는 1) 단순 회귀 분석, 2) Akaike information criteria (AIC) 기반 최적화 회귀 분석 및 3) Random forest (RF)를 활용하였다. 또한, HLS 위성 자료의 적용성을 평가하기 위해 미국 오하이오 주에 위치하고 있는 130여개의 중규모 및 대규모 호소에서 2000년부터 2021년까지 수집된 클로로필-a 관측치를 활용하였다. 두 가지 수질 추정 모형에 대한 정확도 검증에 앞서 오하이오 주 내에서의 클로로필-a의 시계열적 변동성에 대하여 분석하였다. 전반적으로, 2000년부터 2016년까지는 Chl-a가 꾸준히 증가하는 경향성을 나타내었으나, 그 이후로는 감소하는 추세를 나타내었다. 이를 기반으로, 각 방법론을 통해서 나온 Chl-a 추정치에 대해서 통계적 검증을 수행하였다. 결과, 단순 회귀 분석을 통해 추청된 Chl-a값의 결정계수는 0.34였지만, AIC 기반 모델과 RF모형을 사용한 결과 결정계수가 각각 0.82와 0.92로 향상된 것을 확인할 수 있었다. 이와 더불어, spatial 및 temporal window와 더불어 호소의 크기에 따른 정확도 분석 또한 수행하였다. 그 결과, temporal window 가 정확도에 가장 큰 영향을 미치는 것으로 나타났으며, 호소의 크기가 작을수록 정확도가 낮아지는 것을 확인 할 수 있었다. 본 연구의 결과를 토대로 추후 국내 호소에 대해 상기 모형들의 적용성 평가를 수행하여 효율적인 수질 모니터링 시스템 구축으로 이어질 수 있을 것으로 기대된다.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.9
• /
• pp.93-104
• /
• 2023

In modern society, the service method is tended to prefer the non-face-to-face method rather than the face-to-face method. However, services that recommend products such as shoes will inevitably be face-to-face method. In this paper, for the purpose of non-face-to-face service, a system that a foot size is automatically measured and some shoes are recommended based on the measurement result is proposed. To analyze the performance of the proposed method, size measurement error rate and recommendation performance were analyzed. In the recommendation performance experiments, a total of 10 methods for similarity calculation were used and the recommendation method with the best performance among them was applied to the system. From the experiments, the error rate the foot size was small and the recommendation performance was possible to derive significant results. The proposed method is at the laboratory level and needs to be expanded and applied to the real environment. Also, the recommendation method considering design could be needed in the future work.