On the impact of excitation and spectral parameters for expressive statistical parametric speech synthesis.
Ranniery MaiaMasami AkaminePublished in: Comput. Speech Lang. (2014)
Keyphrases
- speech synthesis
- speech recognition
- text to speech
- kernel density estimators
- maximum likelihood
- statistical analysis
- parameter settings
- prosodic features
- statistical distribution
- parameter values
- data sets
- parametric models
- maximum likelihood estimation
- neural network
- statistical information
- spectral analysis
- sensitivity analysis
- hidden markov models
- artificial neural networks
- computer vision