8th European Conference on Speech Communication and Technology

Geneva, Switzerland
September 1-4, 2003


Decision Tree-Based Simultaneous Clustering of Phonetic Contexts, Dimensions, and State Positions for Acoustic Modeling

Heiga Zen, Keiichi Tokuda, Tadashi Kitamura

Nagoya Institute of Technology, Japan

In this paper, a new decision tree-based clustering technique called Phonetic, Dimensional and State Positional Decision Tree (PDS-DT) is proposed. In PDS-DT, phonetic contexts, dimensions and state positions are grouped simultaneously during decision tree construction. PDS-DT provides a complicate distribution sharing structure without any external control parameters. In speaker-independent continuous speech recognition experiments, PDS-DT achieved about 13%-15% error reduction over the phonetic decision tree-based state-tying technique.

Full Paper

Bibliographic reference.  Zen, Heiga / Tokuda, Keiichi / Kitamura, Tadashi (2003): "Decision tree-based simultaneous clustering of phonetic contexts, dimensions, and state positions for acoustic modeling", In EUROSPEECH-2003, 3189-3192.