Practical high-quality speech and voice synthesis using fixed frame rate ABS/OLA sinusoidal modeling.
E. Bryan GeorgePublished in: ICASSP (1998)
Keyphrases
- frame rate
- high quality
- video sequences
- text to speech
- high speed
- spatial resolution
- video quality
- fundamental frequency
- emotion recognition
- d scene
- video camera
- standard pc
- speech recognition
- motion blur
- three dimensional
- motion estimation
- speech synthesis
- multiresolution
- speech sounds
- bitstream
- ground truth
- video rate
- image data
- voice activity detection