Robust Feature Extraction Based on Teager-Entropy and Half Power Spectrum Estimation for Speech Recognition.
Jing DongDongsheng ZhouQiang ZhangPublished in: MIWAI (2015)
Keyphrases
- speech recognition
- power spectrum
- frequency analysis
- frequency domain
- feature extraction
- noisy environments
- speaker identification
- natural images
- hidden markov models
- automatic speech recognition
- pattern recognition
- speech signal
- speech recognizer
- speech processing
- speech synthesis
- language model
- cepstral coefficients
- speech recognition technology
- high order statistics
- speech recognition systems
- speaker independent
- computer vision
- feature space
- feature selection
- palmprint
- noise reduction
- feature vectors
- computational complexity
- image processing
- information retrieval