Noise-Robust Voice Conversion Based on Sparse Spectral Mapping Using Non-negative Matrix Factorization.
Ryo AiharaRyoichi TakashimaTetsuya TakiguchiYasuo ArikiPublished in: IEICE Trans. Inf. Syst. (2014)
Keyphrases
- image noise
- greater robustness
- salt pepper
- rank minimization
- additive noise
- imaging artifacts
- computationally efficient
- noisy environments
- noisy data
- real time face tracking
- noise reduction
- signal to noise ratio
- infrared
- spectral analysis
- geometric distortions
- sparsity constraints
- robust statistical
- natural images
- hidden markov models
- high dimensional
- feature space
- neural network