I am currently an engineer for the Siri Speech team at Apple in Cupertino, CA.
My background is in speech recognition, machine learning, signal processing, and information retrieval.
My past work includes (see publications for details):
- Speaker search: Efficient search of speakers over large populations (10,000 or more).
- Speaker clustering and diarization: Statistical utterance comparison methods for speaker clustering and diarization.
- Music information retrieval: Efficient search of music pitch contours using wavelets and a segmented dynamic time-warping algorithm.
- Auditory modeling for speech processing: Noise robustness in a computational auditory model.
- Category-dependent feature selection: Feature selection from high-dimensional auditory data for speech classification.
Most recently, I've also been working on something totally different, but it's the most amazing and wonderful thing I've ever done. See the picture on the left.
Work e-mail: woojay at apple dot com
Personal e-mail: woojay at gmail dot com