South China Normal University
Overall GPA: 4.06.
B.Eng. in Software Engineering
Speech AI Researcher
HongYu Liu
I study trustworthy speech language models, interactional privacy, and spoken dialogue intelligence, with experience carrying research into production systems.
Education
Academic training in software engineering and data science.
The Chinese University of Hong Kong, Shenzhen
Enrollment scheduled for September 2026.
Master of Science in Data Science
Selected Research
Speech-language-model safety, privacy, and spoken dialogue intelligence.
VoxSafeBench: Not Just What Is Said, but Who, How, and Where
Introduced a benchmark for evaluating social alignment in speech language models across safety, fairness, and privacy dimensions.
VoxPrivacy: A Benchmark for Evaluating Interactional Privacy of Speech Language Models
Introduced a multi-user benchmark for measuring interactional privacy risks in speech-language models.
DialogGraph-LLM: Graph-Informed LLMs for End-to-End Audio Dialogue Intent Recognition
Proposed a graph-informed framework for end-to-end intent recognition in spoken dialogue.
Multi-segment Multitask Fusion Network for Marketing Audio Classification
Proposed MSMT-FN for marketing-call attitude classification, achieving stronger results than prior baselines.
Experience
Research translated into production speech systems.
Speech Algorithm Intern
Developing production-grade speech algorithms and optimizing low-latency model inference for voice applications.
Shenzhen, China
R&D Intern
Developed core algorithms for a voice-cloning application and supported backend model integration for video translation.
Shenzhen, China
Insights
The latest published note from my public research and industry notebook.
A public notebook for analysis that does not fit inside a CV: industry structure, company strategy, and reactions to important reporting.
View all Insights