Sparse and independent representations of speech signals based on parametric models.
Hugo Leonardo RufinerLuís F. RochaJohn Goddard ClosePublished in: INTERSPEECH (2002)
Keyphrases
- parametric models
- speech signal
- speech recognition
- distributed representations
- automatic speech recognition systems
- flow field
- automatic speech recognition
- noisy environments
- hidden markov models
- vocal tract
- background noise
- additive noise
- linear prediction
- blind source separation
- spectral analysis
- sparse representation
- non stationary
- image segmentation
- active appearance models
- sound source
- image acquisition
- speech enhancement
- speech sounds
- pattern recognition