A C/V segmentation algorithm for Mandarin speech signal based on wavelet transforms.
Jhing-Fa WangShi-Huang ChenPublished in: ICASSP (1999)
Keyphrases
- segmentation algorithm
- speech signal
- speech recognition
- wavelet transform
- automatic segmentation
- image segmentation
- automatic speech recognition
- unsupervised segmentation
- region growing
- multiscale
- graph cuts
- speaker identification
- multiresolution
- hidden markov models
- image compression
- language model
- noisy environments
- segmentation method
- pattern recognition
- speech synthesis
- image coding
- superpixels
- automatic speech recognition systems
- non stationary
- high frequency
- subband
- input image
- broadcast news
- wavelet coefficients
- low frequency
- fundamental frequency
- segmentation scheme
- watershed segmentation
- speech quality
- gaussian mixture model
- initial segmentation
- image segmentation algorithm
- feature extraction
- neural network
- machine learning