Feng Ding
Senior Researcher, Machine LearningNRC Beijing
Research Interests
- Mobile context mining, behavior based data mining
- Speech synthesis: Low footprint methods, Unit election methods, Hidden Markov based methods
- Speech recognition: Acoustic Modeling, mainly HMM modeling
- Speech signal processing, Natural language processing, Information extraction, Chinese word segmentation, Part-of-Speech tagging, Dialogue management
Research Projects
- User bevavior data-driven segmentation
- Message dictation
- High Quality Text-To-Speech
- Name dialling
- Low footprint speech synthesis
Personal Information
Please contact me at feng.f.ding@nokia.com or feng.f.ding@gmail.com
Publications
In last three years
- Jingwei Sun; Feng Ding; Yahui Wu, A polynomial segment model based statistical parametric speech synthesis system, 2009 IEEE International Conference on Acoustics, Speech and Signal Processing
- Feng Ding, Jani Nurminen, Jilei Tian, Efficient Join Cost Computation for Unit Selection based TTS Systems, 9th Annual Conference of the International Speech Communication Association , Interspeech 2008, Brisbane.
- Feng Ding, Jari Alhonen, NTTS participation in the Blizzard Challenge 2008, Blizzard workshop 2008, Brisbane.
- Feng Ding, Jari Alhonen, Non-uniform Unit Selection through Search Strategy for Blizzard Challenge 2007, Blizzard workshop 2007, Bonn.
Patents
- Tone estimation for name
- High quality Thai text-to-phoneme converter
