SVM-based separation of unvoiced-voiced speech in cochannel conditions.
Ke HuDeLiang WangPublished in: ICASSP (2012)
Keyphrases
- speech signal
- speech recognition
- sound source
- automatic speech recognition
- speech quality
- linear prediction
- vocal tract
- hidden markov models
- speech processing
- speaker identification
- background noise
- noisy environments
- automatic speech recognition systems
- support vector
- non stationary
- emotion recognition
- spectral analysis
- blind source separation
- speech synthesis
- acoustic features
- support vector machine
- pattern recognition
- support vector machine svm
- speaker recognition
- relevance feedback
- active learning
- data sets
- feature set
- information retrieval