This year (2025), I joined Dolby as a Senior Foundational AI Researcher, based in Atlanta.
I'm finishing Ph.D. degree in Intelligent System Engineering at Indiana University Bloomington, advised by Professor Minje Kim. My research interest is in understanding high-dimensional audio data with effective and efficient representation for solving the audio related problems. My research work during Ph.D. include:
- Audio Coding
- - Source-aware audio codec
- - Low-bitrate predictive audio codec
- - Latent diffusion-based low-bitrate audio codec
- Generative Modeling on Speech and Audio
- - Auto-regressive generative speech enhancement, with discrete or continuous features
- - A survey on Discrete audio tokenizers
- Music Applications
- - Music remixing, with controling in the latent space;
- - Music upmixing, from stereo to 5-channel.