• Title/Summary/Keyword: STT

Search Result 172, Processing Time 0.022 seconds

LLM-based chatbot system to improve worker efficiency and prevent safety incidents (작업자의 업무 능률 향상과 안전 사고 방지를 위한 LLM 기반 챗봇 시스템)

  • Doohwan Kim;Yohan Han;Inhyuk Jeong;Yeongseok Hwnag;Jinju Park;Nahyeon Lee;Yujin Lee
    • Proceedings of the Korean Society of Computer Information Conference
    • /
    • 2024.01a
    • /
    • pp.321-324
    • /
    • 2024
  • 본 논문에서는 LLM(Large Language Models) 기반의 STT 결합 챗봇 시스템을 제안한다. 제조업 공장에서 안전 교육의 부족과 외국인 근로자의 증가는 안전을 중시하는 작업 환경에서 새로운 도전과제로 부상하고 있다. 이에 본 연구는 언어 모델과 음성 인식(Speech-to-Text, STT) 기술을 활용한 혁신적인 챗봇 시스템을 통해 이러한 문제를 해결하고자 한다. 제안된 시스템은 작업자들이 장비 사용 매뉴얼 및 안전 지침을 쉽게 접근하도록 지원하며, 비상 상황에서 신속하고 정확한 대응을 가능하게 한다. 연구 과정에서 LLM은 작업자의 의도를 파악하고, STT 기술은 음성 명령을 효과적으로 처리한다. 실험 결과, 이 시스템은 작업자의 업무 효율성을 증대시키고 언어 장벽을 해소하는데 효과적임이 확인되었다. 본 연구는 제조업 현장에서 작업자의 안전과 업무 효율성 향상에 기여할 것으로 기대된다.

  • PDF

The Impact of Speech-To-Text-based Class on Learners' Cognitive Abilities

  • HyunMin Kang;SunKwan Han
    • Journal of the Korea Society of Computer and Information
    • /
    • v.29 no.1
    • /
    • pp.287-293
    • /
    • 2024
  • This research studied the cognitive impact of classes using artificial intelligence on aviation technical school students. First, we developed a class consisting of a class based on traditional presentation materials and a class composed of speech-to-text (STT)-based artificial intelligence materials. A 133 students from an aviation education institution participated in two types of classes. We measured students' cognitive load and Mind Wandering test results before and after class, and conducted an achievement evaluation. As a result of the test analysis, we confirmed that extraneous cognitive load was reduced, content concentration increased, and achievement improved. In the future, we hope that AI-based STT classes will be widely used in schools that teach technology.

Analysis and Improvement of Time Sampling effects on Singular Perturbation based Control Systems - Its Aplication to Design of Singular Pertubation based STT Missible Digital Autopilot (특이섭동 기법 기반 제어 시스템에 대한 샘플링 영향 분석 및 개선 - 특이섭동 기법 기반 STT 미사일 디지털 자동조정장치 설계에의 적용)

  • Jeong, Seon-Tae
    • Journal of the Institute of Electronics Engineers of Korea SC
    • /
    • v.37 no.3
    • /
    • pp.33-43
    • /
    • 2000
  • The guarantee of the fast dynamics stability is essential for successful application of singular Perturbation technique to control systems design. Even though the fast dynamics of the control systems is rendered stable by an analog controller, the fast dynamics stability of the control systems resulted from an digital implementation of the analog controller can be impaired severely. In this paper, we first investigate the time sampling effects on singular perturbation based control systems by centering on a design example of recently developed singular perturbation based STT missile autopilot with high performance. The investigation shows that the stability margin the fast dynamics of the STT misile autopilot system decreases rapidly as the sampling interval of discretizing the analog autopilot increases. Under this analysis, we propose a composite digital controller with compensation for the decreasing stability margin of the fast dynamics due to time sampling to achieve better performance with respect to sampling time. The improved performance of the proposed composite digital controller is verified by simulation. This result shows that one needs to investigate time sampling effects in the digital implementation of singular perturbation based controllder, and then can have benefit from the investigation.

  • PDF

A Design and Implementation of Speech Recognition and Synthetic Application for Hearing-Impairment

  • Kim, Woo-Lin;Ham, Hye-Won;Yun, Sang-Un;Lee, Won Joo
    • Journal of the Korea Society of Computer and Information
    • /
    • v.26 no.12
    • /
    • pp.105-110
    • /
    • 2021
  • In this paper, we design and implement an Android mobile application that helps hearing impaired people communicate based on STT(Speech-to-Text) and TTS(Text-to-Speech) APIs and accelerometer sensor of a smartphone. This application provides the ability to record what the hearing-Impairment person's interlocutor is saying with a microphone, convert it to text using the STT API, and display it to the hearing-Impairment person. In addition. In addition, when a hearing-impaired person inputs a text using the TTS API, it is converted into voice and told to the interlocutor. When a hearing-impaired person shakes their smartphone, an accelerometer based background service function is provided to run the application. The application implemented in this paper provides a function that allows hearing impaired people to communicate easily with other people when communicating with others without using sign language as a video call.

Association of serum 25-hydroxyvitamin D concentrations with Schirmer tear test 1 and tear film breakup time in dogs

  • Youngsam Kim;Seonmi Kang;Kangmoon Seo
    • Journal of Veterinary Science
    • /
    • v.24 no.2
    • /
    • pp.32.1-32.10
    • /
    • 2023
  • Background: The association between vitamin D and canine keratoconjunctivitis sicca (KCS) has not been investigated in dogs. Objectives: To investigate the association of serum 25-hydroxyvitamin D [25(OH)D] concentrations with Schirmer tear test 1 (STT-1) and tear film breakup time (TFBUT) in dogs. Methods: Sixty-one clinically healthy, client-owned dogs were enrolled. STT-1 and TFBUT were measured in 122 (61dogs) and 82 (41 dogs out of total 61 dogs) eyes, respectively. Serum 25(OH)D concentrations were evaluated by quantitative chemiluminescent immunoassay. The dogs were classified into 6 groups according to the evaluations (STT-1: group 1, normal [≥ 15 mm/min] in both eyes; group 2, normal in one eye and abnormal [< 15 mm/min] in the fellow eye; group 3, abnormal in both eyes; TFBUT: group 4, normal [≥ 20 sec] in both eyes; group 5, normal in one eye and abnormal [< 20 sec] in the fellow eye; group 6, abnormal in both eyes). Results: STT-1 was positively correlated with TFBUT (p < 0.001). Among the STT-1 groups, the mean serum 25(OH)D concentration in group 1 was significantly higher than in groups 2 and 3 with positive correlation (p < 0.001). However, there were no significant differences among the TFBUT groups 4, 5, and 6. Conclusions: In dogs, it was found that serum 25(OH)D concentrations had a greater effect on quantitative KCS than qualitative KCS. Therefore, it is considered that measurement of serum 25(OH)D concentration could be included in the diagnostic tests in canine quantitative KCS patients.

Dose Distribution and Design of Dynamic Wedge Filter for 3D Conformal Radiotherapy (방사선 입체조형치료를 위한 동적쐐기여과판의 고안과 조직내 선량분포 특성)

  • 추성실
    • Progress in Medical Physics
    • /
    • v.9 no.2
    • /
    • pp.77-88
    • /
    • 1998
  • Wedge shaped isodoses are desired in a number of clinical situations. Hard wedge filters have provided nominal angled isodoses with dosimetric consequences of beam hardening, increased peripheral dosing, nonidealized gradients at deep depths along with the practical consequendes of filter handling and placement problems. Dynamic wedging uses a combination of a moving collimator and changing monitor dose to achieve angled isodoses. The segmented treatment tables(STT) that monitor unit setting by every distance of moving collimator, was induced by numerical formular. The characteristics of dynamic wedge by STT compared with real dosimetry. Methods and Materials : The accelerator CLINAC 2100C/D at Yonsei Cancer Center has two photon energies (6MV and 10MV), currently with dynamic wedge angles of 15$^{\circ}$, 30$^{\circ}$, 45$^{\circ}$ and 60$^{\circ}$. The segmented treatment tables(STT) that drive the collimator in concert with a changing monitor unit are unique for field sizes ranging from 4.0cm to 20.0cm in 0.5cm steps. Transmission wedge factors were measured for each STT with an standard ion chamber. Isodose profiles, isodose curves, percentage depth dose for dynamic wedge filters were measured with film dosimetry. Dynamic wedge angle by STT was well coincident with film dosimetry. Percent depth doses were found to be closer to open field but more shallow than hard wedge filter. The wedge transmission factor were decreased by increased the wedge angle and more higher than hard wedge filters. Dynamic wedging probided more consistent gradients across the field compared with hard wedge filters. Dynamic wedging has practical and dosimetric advantages over hard filters for rapid setup and keeping from table collisions. Dynamic wedge filters are positive replacement for hard filters and introduction of dynamic conformal radiotherapy and intensity modulation radiotherapy in a future.

  • PDF

STT-MRAM Read-circuit with Improved Offset Cancellation

  • Lee, Dong-Gi;Park, Sang-Gyu
    • JSTS:Journal of Semiconductor Technology and Science
    • /
    • v.17 no.3
    • /
    • pp.347-353
    • /
    • 2017
  • We present a STT-MRAM read-circuit which mitigates the performance degradation caused by offsets from device mismatches. In the circuit, a single current source supplies read-current to both the data and the reference cells sequentially eliminating potential mismatches. Furthermore, an offset-free pre-amplification using a capacitor storing the mismatch information is employed to lessen the effect of the comparator offset. The proposed circuit was implemented using a 130-nm CMOS technology and Monte Carlo simulations of the circuit demonstrate its effectiveness in suppressing the effect of device mismatch.

Low-Power Write-Circuit with Status-Detection for STT-MRAM

  • Shin, Kwang-Seob;Im, Saemin;Park, Sang-Gyu
    • JSTS:Journal of Semiconductor Technology and Science
    • /
    • v.16 no.1
    • /
    • pp.23-30
    • /
    • 2016
  • We report a STT-MRAM write-scheme, in which the length of the write-pulse is determined dynamically by sensing the status of MTJ cells. The proposed scheme can reduce the power consumption by eliminating unnecessary writing current after the switching has occurred. We also propose a reference cell design, which is optimized for the use in write-circuits. The performance of the proposed circuit was verified by SPICE level simulations of the circuit implemented in a $0.13{\mu}m$ CMOS process.

Subtitle Automatic Generation System using Speech to Text (음성인식을 이용한 자막 자동생성 시스템)

  • Son, Won-Seob;Kim, Eung-Kon
    • The Journal of the Korea institute of electronic communication sciences
    • /
    • v.16 no.1
    • /
    • pp.81-88
    • /
    • 2021
  • Recently, many videos such as online lecture videos caused by COVID-19 have been generated. However, due to the limitation of working hours and lack of cost, they are only a part of the videos with subtitles. It is emerging as an obstructive factor in the acquisition of information by deaf. In this paper, we try to develop a system that automatically generates subtitles using voice recognition and generates subtitles by separating sentences using the ending and time to reduce the time and labor required for subtitle generation.

AI voice phishing prevention solution using Open STT API and machine learning (Open STT API와 머신러닝을 이용한 AI 보이스피싱 예방 솔루션)

  • Mo, Shi-eun;Yang, Hye-in;Cho, Eun-bi;Yoon, Jong-Ho
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2022.11a
    • /
    • pp.1013-1015
    • /
    • 2022
  • 본 논문은 보이스피싱에 취약한 VoIP와 일반 유선전화 상의 보안을 위해 유선전화의 대화내용을 Google STT API 및 텍스트 자연어 처리를 통해 실시간으로 보이스피싱 위험도를 알 수 있는 모델을 제안했다. 보이스피싱 데이터를 Data Augmentation와 BERT 모델을 활용해 보이스피싱을 예방하는 솔루션을 구상했다.