Discrete acoustic space for an efficient sampling in neural text-to-speech.
Marek StrelecJonas RohnkeAntonio BonafonteMateusz LajszczakTrevor WoodPublished in: CoRR (2021)
Keyphrases
- text to speech
- prosodic features
- discrete space
- speech synthesis
- continuous space
- neural network
- space time
- parameter space
- word processing
- underwater vehicles
- network architecture
- higher dimensional
- text to speech synthesis
- programming tool
- sample space
- english text
- search space
- sampling algorithm
- learning rules
- random sampling
- source localization
- multi modal
- high dimensional
- pattern recognition