自訂 Cookie
禁止且拒絕未經各資訊當事人同意，擅自蒐集本服務提供的使用者個人資訊資料等資料之行為。即使是公開資料，若未經許可使用爬蟲等技術裝置進行蒐集，依個人資訊保護法可能會受到刑事處分，特此告知。
© 2025 Rocketpunch, 주식회사 더블에이스, 김인기, 大韓民國首爾特別市城東區聖水一路10街 12, 12樓 1號, 04793, support@rocketpunch.com, +82 10-2710-7121
統一編號 206-87-09615
更多
IsmoilovNusrat
I'm interested in Prompt Engineering, LLMs, STT, TTS, NLP, VAD, and Speaker Diarization.
經歷
**Built Voice Conversational Assistant Application for Mobile Devices which works via Smartwatches over Bluetooth. Used GPT-3/3.5, Naver HyperClova, and Google Speech Services. **Built an application for Keyword Spotting with TC-ResNet model for Korean words for Android Mobile Devices. **Built an application for Speech To Text with Conformer for Android Mobile Devices.
更多
學歷
活動
最近活動
獲獎 1
證照 1
專案 6
프로젝트
• Designed and Developed a Voice Assistant Application
(주) 키핀 · 2022년 10월 - 2023년 9월 · 1년
Served as the sole engineer for a voice assistant that operates on Android mobile phones via Bluetooth Classic from a smartwatch. The application includes the following features: o Created a daily recommendation engine powered by GPT-4 that processes sleep data (covering all sleep stages like deep
프로젝트
• Keyword Spotting Application in Noisy Environment for Android mobile phones
(주) 키핀 · 2022년 1월 - 2022년 5월 · 5개월
o Designed, trained, and deployed a lightweight background noise classification model for audio recordings on Android applications. (Technologies: TensorFlow Inference, TensorFlow Lite) o Deployed a lightweight model for Keyword Spotting (KWS) of Korean words on Android applications. (Technologies:
자격증
Web Application Technologies and Django
2021년 2월
Basic of the Web Application in Django
프로젝트
Extracting HB(Heart Beat) artifact from EEG dataset
2019년 10월 - 2020년 1월 · 4개월
I developed a CNN model for deleting/extracting HB noise from EEG data. As the data, I used PSD (Power Spectrum Density) graph images of HB.
프로젝트
Develop Super Resolution model and implement on FFMPEG
2019년 5월 - 2019년 9월 · 5개월
I developed a light-weight (SRCNN) as well as heavy-weight (WDSR, EDSR) for Super Resolution task. Prepared the trained models for deployment via FFMPEG (core Tensorflow)