Enhancement of noisy speech recordings via blind source separation.
Jirí MálekZbynek KoldovskýJindrich ZdánskýJan NouzaPublished in: INTERSPEECH (2008)
Keyphrases
- blind source separation
- speech signal
- speech enhancement
- independent component analysis
- single channel
- acoustic features
- non stationary
- speech recognition
- image enhancement
- image processing
- negative matrix factorization
- background noise
- noisy environments
- signal processing
- factor analysis
- audio visual
- sparse representation
- automatic speech recognition
- spectral analysis
- hidden markov models
- pattern recognition
- contrast enhancement
- multi channel