Combination of Recognizers and Fusion of Features Approach to Missing Data ASR Under Non-Stationary Noise Conditions.
Neil JoshiLing GuanPublished in: ICASSP (4) (2007)
Keyphrases
- missing data
- non stationary
- missing values
- motion segmentation
- structure from motion
- low rank
- incomplete data
- noisy data
- missing information
- adaptive algorithms
- speech signal
- automatic speech recognition
- white noise
- noisy environments
- empirical mode decomposition
- imprecise data
- data imputation
- matrix factorization
- feature space
- feature extraction
- extracted features
- image denoising
- em algorithm
- expectation maximization
- hidden markov models
- computer vision
- database