A new method of generating speech synthesis units based on phonological knowledge and clustering technique.
Yuki YoshidaShin'ya NakajimaKazuo HakodaTomohisa HirokawaPublished in: ICSLP (1996)
Keyphrases
- clustering method
- prior knowledge
- k means
- hierarchical clustering
- preprocessing
- pairwise
- synthetic data
- unsupervised learning
- unsupervised clustering
- similarity function
- high accuracy
- knowledge management
- input data
- speech synthesis
- data sets
- knowledge acquisition
- domain knowledge
- significant improvement
- pattern recognition
- image segmentation
- clustering algorithm