Residual compensation based on articulatory feature-based phone clustering for hybrid Mandarin speech synthesis.
Yi-Chin HuangChung-Hsien WuShih-Lun LinPublished in: SSW (2013)
Keyphrases
- speech synthesis
- speech recognition
- prosodic features
- vocal tract
- text to speech
- hidden markov models
- clustering method
- clustering algorithm
- pattern recognition
- automatic speech recognition
- speech corpus
- data clustering
- language model
- information theoretic
- cluster analysis
- speech recognizer
- speech signal
- unsupervised learning
- data points
- noisy environments
- data sets
- high dimensional data
- k means
- probabilistic model
- hybrid learning
- data mining tasks
- fuzzy c means
- fuzzy clustering
- hierarchical clustering