C/V Segmentation on Mandarin Spontaneous Spoken Speech Signals Using SNR Improvement and Energy Variation.
Ching-Ta LuHsiao-Chuan WangPublished in: IEICE Trans. Inf. Syst. (2006)
Keyphrases
- speech recognition
- speech signal
- automatic speech recognition
- speech enhancement
- conversational speech
- broadcast news
- hidden markov models
- level set
- signal to noise ratio
- linear prediction
- speaker identification
- language model
- noisy environments
- pattern recognition
- automatic speech recognition systems
- blind source separation
- spectral analysis
- speech synthesis
- vocal tract
- background noise
- non stationary
- image segmentation
- speech sounds
- sound signals
- additive noise
- word recognition
- noise reduction
- noisy images
- multiscale