Autoregressive Variational Autoencoder with a Hidden Semi-Markov Model-Based Structured Attention for Speech Synthesis.
Takato FujimotoKei HashimotoYoshihiko NankakuKeiichi TokudaPublished in: ICASSP (2022)
Keyphrases
- autoregressive
- semi markov
- speech synthesis
- speech recognition
- random fields
- conditional random fields
- non stationary
- text to speech
- moving average
- vocal tract
- gaussian markov random field
- autoregressive model
- sar images
- markov random field
- linear prediction
- probabilistic model
- hidden markov models
- speech signal
- autoregressive moving average
- pattern recognition
- image segmentation
- neural network
- parameter estimation
- arma model
- spectrum analysis
- random field models
- information extraction
- higher order
- image denoising