A Unified Trajectory Tiling Approach to High Quality Speech Rendering.
Yao QianFrank K. SoongZhi-Jie YanPublished in: IEEE Trans. Speech Audio Process. (2013)
Keyphrases
- high quality
- high fidelity
- speech recognition
- ground truth
- low quality
- speech signal
- higher quality
- high resolution
- speech synthesis
- trajectory data
- image quality
- unified model
- computer graphics
- text to speech
- recognition engine
- endpoint detection
- real time
- highly accurate
- computer vision
- audio visual
- super resolution
- volume rendering
- data sets
- angular velocity
- spontaneous speech
- mobile robot
- multi stream
- speaker recognition
- spoken language
- graphics hardware
- single image
- depth map
- noisy environments
- automatic speech recognition