Feature Extraction with Combination of HMT-Based Denoising and Weighted Filter Bank Analysis for Robust Speech Recognition.
Sungyun JungJong-Mok SonKeun-Sung BaePublished in: IEICE Trans. Inf. Syst. (2005)
Keyphrases
- speech recognition
- filter bank
- denoising
- wavelet domain
- noisy images
- subband
- feature extraction
- hidden markov models
- pattern recognition
- wavelet packet
- computationally efficient
- image coding
- image denoising
- wavelet transform
- language model
- spectral analysis
- multiscale
- frequency domain
- speaker identification
- multiresolution
- face recognition
- image processing
- em algorithm
- feature space
- computational complexity
- high quality
- speech signal
- feature selection
- computer vision