Harmonics Plus Noise Model Based Vocoder for Statistical Parametric Speech Synthesis.
Daniel ErroIñaki SainzEva NavasInma HernáezPublished in: IEEE J. Sel. Top. Signal Process. (2014)
Keyphrases
- speech synthesis
- speech recognition
- text to speech
- data driven
- vocal tract
- noisy data
- noisy environments
- statistical analysis
- prosodic features
- noise level
- noise reduction
- gaussian noise
- statistical information
- additive noise
- statistical models
- noise sensitivity
- information theoretic
- computer vision
- median filter
- model free
- fully unsupervised
- high frequency
- frequency domain
- hidden markov models
- multiresolution