CV

So Yeong Park · 박소영
Undergraduate Researcher, AINC Lab, Hanshin University
kong02931@gmail.com · GitHub

Research interests

Korean speech recognition; Audio-language understanding and retrieval; Reliable evaluation of language models.

Hanshin University
School of AI & Software · Mar 2024 – H1 2027 (expected)
Bachelor’s degree in progress. Double major in AI-Integrated Disability Life Care.
Cumulative GPA: 4.38 / 4.50 · 96 credits completed.

Undergraduate Researcher, AINC Lab
Hanshin University · Jun 2025 – Present
Advisor: Prof. Seon Man Kim
Research design, experiment implementation, results analysis, and paper writing.

2026

Compared Qwen3-ASR 0.6B and 1.7B on 10 Korean evaluation sets.
Analyzed language settings, maximum generation length, temperature, and repetition penalty using character error rate.
Designed the study, implemented experiments, analyzed results, and wrote the paper.

2026 — Present

Reproduced the official baseline and its full evaluation pipeline.
Experimenting with MS-CLAP audio-language representations and DETR-based temporal localization models.

2025 — 2026

Designed a rubric-based dataset for academic-regulation questions and answers in 2025.
Designed and reviewed 9,958 synthetic examples labeled PASS, FAIL, and OOS; fine-tuned Qwen2.5-7B-Instruct and Llama-3.1-8B with 4-bit QLoRA.
Measured a FAIL correction rate of 0.9897 when ground-truth review signals were supplied; identified error review as a bottleneck in end-to-end inference.

So Yeong Park, Seon Man Kim. Evaluation and inference parameter analysis of Qwen3-ASR on multiple Korean speech corpora. The Journal of the Acoustical Society of Korea, 45(5), 582–590, 2026.
First author · Accepted Aug 25, 2026.

Hyeon Jae Shin, Jin Tae Seok, So Yeong Park, Seon Man Kim. Low-Latency Deep Learning-Based Denoising of Self-Generated Footstep Noise in Quadruped Robots for Remote Acoustic Situational Awareness. 16th International Conference on Theoretical and Computational Acoustics (ICTCA 2025), Aug 2025 · Busan.
Third author.

So Yeong Park, Seon Man Kim. A Correction-Aware Synthetic Dataset for Training Language Models in Academic-Regulation Question Answering. Korean Association of Data Science Summer Conference, Jun 19, 2026.
First author · Paper Encouragement Award.
So Yeong Park, Hyeon Jae Shin, Seon Man Kim. Inference Parameter Optimization of Qwen3-ASR across Multiple Korean Speech Corpora. Institute of Electronics and Information Engineers Summer Conference · P4-80, 2026.
First author.
So Yeong Park, Seon Man Kim. Inference Parameter Optimization of a Public Multilingual ASR Model for Korean Speech Recognition. Joint Spring Conference · Korean Society of Speech Sciences & Phonology-Morphology Circle of Korea, Jun 13, 2026.
First author.
So Yeong Park, Min Jun Kim, Seon Man Kim. Design and Construction of a Rubric-Based Training Dataset for University Academic Regulations. Korean Institute of Smart Media Fall Conference, Nov 15, 2025.
First author · Outstanding Paper Award.

Paper Encouragement Award, Korean Association of Data Science (Jun 2026).
Synthetic dataset for error review and correction training in academic-regulation QA
Outstanding Paper Award, Korean Institute of Smart Media (Nov 2025).
Rubric-based training dataset for university academic regulations
Grand Prize, Hanshin University (Nov 2025).
Hanshin University AI Festival, General Division · Event coverage
Top Excellence Award, SW-Centered University Council (Oct 2025).
2025 Elysian Gangchon AI Hackathon · Event coverage
Outstanding Paper Award, DAEHAN Society of Industrial Management (Aug 2025).
Improving service delivery for AI-based assistive devices and welfare equipment for people with disabilities and older adults
Grand Prize, Korea University SW-Centered University Program (May 2025).
AI/Blockchain Startup Competition

Academic Excellence Scholarship, Hanshin University (During studies).
Awarded every semester for academic excellence.
ILJU Academy and Culture Foundation, Taekwang Group (Jul 2025 — Graduation).
Selected for the 33rd cohort of scholars.

Teaching Assistant (Spring & Fall 2026).
Supported Mathematics for AI & SW and Applied Deep Learning courses.

AIUX Lab Leader (Jun 2026 — Present).
Organize regular activities and research exchange for an undergraduate research community.
Volunteer Service (400+ hours in total).
More than 400 cumulative volunteer hours across activities. I tutor elementary school students weekly as an ILJU scholar.

CES 2026 Participant (Jan 2026).
Selected by Hanshin University to attend CES 2026.

Updated October 2026