FlexVoice: A Parametric Approach to High-Quality Speech Synthesis.
György BaloghErvin DoblerTamás GröblerBéla SmodicsCsaba SzepesváriPublished in: TSD (2000)
Keyphrases
- speech synthesis
- high quality
- speech recognition
- text to speech
- vocal tract
- prosodic features
- higher quality
- low quality
- ground truth
- speech corpus
- data sets
- high resolution
- image quality
- shape grammars
- high fidelity
- hidden markov models
- evolutionary algorithm
- image processing
- parametric models
- real time
- language model
- similarity measure
- machine learning
- neural network
- yields high quality