Waveform Generation Based on Signal Reshaping for Statistical Parametric Speech Synthesis.
Felipe EspicCassia Valentini-BotinhaoZhizheng WuSimon KingPublished in: INTERSPEECH (2016)
Keyphrases
- speech synthesis
- speech recognition
- text to speech
- frequency domain
- prosodic features
- vocal tract
- signal processing
- high frequency
- frequency spectrum
- non stationary
- statistical analysis
- data driven
- machine learning
- low frequency
- fundamental frequency
- information theoretic
- image acquisition
- hidden markov models
- pattern recognition