I am an undergraduate from Shanghai Jiao Tong University, School of Mechanical Engineering, advised by Prof. Chen Xie at the X-LANCE Lab.

I am also the founder and CEO of Shanghai Xhigh Intelligence Technology Co., Ltd. (上海交辉数智科技有限公司), founded on 18 June 2026. My work centers on voice agents — end-to-end, low-latency speech interaction — and on the agent-native education products built on top of them.

🎓 Education

• 2022.09 - present, Shanghai Jiao Tong University, School of Mechanical Engineering, Shanghai, B.S.
• 2025.09 - 2026.01, National Taiwan University, Department of Electrical Engineering, Taiwan, Intern

🏢 Entrepreneurship

• 2026.06 - present, Founder & CEO, Shanghai Xhigh Intelligence Technology Co., Ltd., Shanghai
  • An Agent × Education company, founded 2026.06.18 and run as a one-person company (OPC): one human founder plus a standing team of AI agents covering R&D, content, data and operations.
  • Voice agents are the core technical direction — end-to-end low-latency speech streams with synchronized visual generation, built on our regional speech–language modeling work.
  • Flagship platform sjtuagent.com is live at SJTU across 7 courses; the same engine extends to primary-school mathematics (school.sjtuagent.com) and children’s AI creation (easy-vibecoding.com).
  • Technically advised by the SJTU Network & Information Center and the SJTU Student Innovation Center; sponsored by the SJTU X-LANCE Lab.

📝 Publications & Patents

• Hao-Hui Xie, Ho-Lam Chung, Yi-Cheng Lin, Ke-Han Lu, Wenze Ren, Xie Chen, and Hung-yi Lee. “TW-Sound580K: A Regional Audio-Text Dataset with Verification-Guided Curation for Localized Audio-Language Modeling.” Under Review, ISCSLP 2026, arXiv:2603.05094. [Paper]
• Tengjie Zhu, Guanyu Cai, Yang Zhaohui, Guanzhu Ren, Hao-Hui Xie, ZiRui Wang, Junsong Wu, Jingbo Wang, Xiaokang Yang, Yao Mu, and Yichao Yan. “CLOT: Closed-Loop Global Motion Tracking for Whole-Body Humanoid Teleoperation.” Under Review, CoRL 2026, arXiv:2602.15060. [Paper]

🚀 Projects

• sjtuagent.com — Turning every course into an agent that can teach. A multi-course teaching platform live at SJTU: 7 courses, 48 chapters and 954 practice questions, offering an AI course tutor (RAG), randomized quizzes, a real-time voice tutor (streaming ASR + TTS), force-directed knowledge graphs, per-concept animations and mock exams. Since launch: 19,900+ learning events, 7,400+ quiz attempts, 1,900+ study sessions and 640+ tutor Q&A turns. Next.js 15 + Python voice backend.
• clawsjtu.com — OpenClaw SJTU Square, an agent-only social square. Every SJTU student raises an “OpenClaw lobster” that posts, comments and argues exclusively through a v1 REST API — humans can only watch. 110 registered agents, gated to @sjtu.edu.cn, with agent breeding and genetics, weekly matchmaking, an XP economy and an MBTI benchmark. Next.js 15 + Supabase.
• openclaw-sjtu (57★) — An AI campus assistant for SJTU students, built on the OpenClaw skill framework. Covers 21+ features including homework tracking, course reviews, Shuiyuan community summarization, and a PPT generator with SJTU templates.

🏅 Honors & Awards

• 2023.02 Mathematical Contest in Modeling (MCM), Award M
• 2023.12 SJTU Academic Excellence Scholarship (Category C)
• 2024.12 SJTU Academic Excellence Scholarship (Category B)
• 2024.05 SJTU Outstanding Student Leader

📌 Academic Conferences

Attended the 17th Asian Conference on Machine Learning (ACML 2025) in Taipei, Taiwan.

💻 Research Experience

• 2025.04 - present, ScaleLab at Shanghai Jiao Tong University, Advisor: Yao Mark Mu
• 2025.07 - 2025.08, Zhejiang Lingqiao Intelligent Technology Co., Ltd., AI & Embedded Systems Intern
• 2025.09 - 2026.01, Speech Processing Lab, National Taiwan University, Advisor: Hung-yi Lee
  • Fine-tuned large-scale Taiwanese speech–language models using the DeSTA2.5 framework, integrating Llama3-8B as the text backbone and Whisper-v3 as the acoustic encoder. Implemented LoRA-based alignment and timestamp-aware fine-tuning to enhance accent adaptation and improve ASR robustness across diverse regional dialects.
  • Designed and optimized a full multi-GPU training and evaluation pipeline, including data preprocessing, feature extraction, LoRA parameter-efficient training, and distributed evaluation across speech benchmarks such as TAU-TW Benchmark. Achieved a relative improvement of over 8–10% in recognition accuracy compared to baseline models.
  • Developed and experimented with speech–language alignment strategies that combine contrastive loss and multi-level feature fusion, enabling the model to perform cross-lingual reasoning and contextually grounded transcription under complex acoustic variations (e.g., Taiwanese Mandarin, Hokkien, and mixed-accent speech).
• 2026.01 - present, X-LANCE Lab, Shanghai Jiao Tong University, Advisor: Chen Xie

📰 Media

• 2026.05.19, SJTU HPC Center (“交我算”) featured openclaw-sjtu as an outstanding case of the Ziyuan-1 cluster program — an all-round AI campus assistant for SJTU students. [Report]