Segment selection method based on tonal validity evaluation using machine learning for concatenative speech synthesis.
Akihiro YoshidaHideyuki MizunoKazunori ManoPublished in: ICASSP (2008)
Keyphrases
- machine learning
- evaluation method
- high accuracy
- preprocessing
- significant improvement
- high precision
- model selection
- speech synthesis
- gold standard
- selection algorithm
- machine learning methods
- classification method
- support vector machine svm
- information extraction
- support vector machine
- probabilistic model
- experimental evaluation
- cost function
- edge detection
- clustering method
- synthetic data
- speech recognition
- segmentation method
- classification accuracy
- evolutionary algorithm
- pattern recognition
- fine tuning
- face recognition