Minimum generation error training with direct log spectral distortion on LSPs for HMM-based speech synthesis.
Yi-Jian WuKeiichi TokudaPublished in: INTERSPEECH (2008)
Keyphrases
- speech synthesis
- speech recognition
- hidden markov models
- text to speech
- error propagation
- vocal tract
- training set
- error rate
- prosodic features
- training samples
- square error
- test set
- quantization error
- error bounds
- spectral analysis
- generation process
- training process
- prediction error
- training examples
- speech corpus
- reconstruction error
- multispectral images
- training phase
- error analysis
- machine learning
- minimum cost
- supervised learning
- image processing