Integration of speaker and pitch adaptive training for HMM-based singing voice synthesis.
Kanako ShirotaKazuhiro NakamuraKei HashimotoKeiichiro OuraYoshihiko NankakuKeiichi TokudaPublished in: ICASSP (2014)
Keyphrases
- acoustic features
- hidden markov models
- data fusion
- cmac neural network
- training process
- speech recognition
- test set
- training examples
- audio visual
- training set
- program synthesis
- mel frequency cepstral coefficients
- data integration
- training samples
- multi modal
- automatic speech recognition
- music information retrieval
- prosodic features
- databases