Core areas
- Noise reduction and source separation using modern neural architectures
- Keyword spotting, ASR, and speaker recognition
- Training, inference, and deployment optimization
- Embedded systems, NPU integration, and fixed-point quantization
I build speech and audio systems that move from research ideas into reliable embedded products.
I graduated from East China Normal University with a master's degree in Computer Technology in 2015, and I have worked at UvoiceTech for more than 10 years, with a focus on embedded-side audio processing and deep-learning-driven product improvement.
Work style
- Move from prototype to production with a strong systems mindset and practical neural-network design
- Balance model quality, latency, memory, and hardware constraints
- Translate algorithm ideas into deployable, maintainable implementations