Estimation of vocal-tract shape from speech spectrum and speech resynthesis based on a generative model.
Tokihiko KaburagiPublished in: INTERSPEECH (2014)
Keyphrases
- vocal tract
- generative model
- speech signal
- speech synthesis
- image acquisition
- speech recognition
- probabilistic model
- speech enhancement
- speech sounds
- linear prediction
- text to speech
- em algorithm
- language model
- automatic speech recognition
- topic models
- noisy environments
- hidden markov models
- prior knowledge
- non stationary
- expectation maximization
- pattern recognition