8th European Conference on Speech Communication and Technology

Geneva, Switzerland
September 1-4, 2003


A Memory-Based Approach to Cantonese Tone Recognition

Michael Emonts (1), Deryle Lonsdale (2)

(1) Sony Electronics, USA
(2) Brigham Young University, USA

This paper introduces memory-based learning as a viable approach for Cantonese tone recognition. The memory-based learning algorithm employed here outperforms other documented current approaches for this problem, which is based on neural networks. Various numbers of tones and features are modeled to find the best method for feature selection and extraction. To further optimize this approach, experiments are performed to isolate the best feature weighting method, the best class voting weights method, and the best number of k-values to implement. Results and possible future work are discussed.

Full Paper

Bibliographic reference.  Emonts, Michael / Lonsdale, Deryle (2003): "A memory-based approach to Cantonese tone recognition", In EUROSPEECH-2003, 2305-2308.