Blind upmix of stereo music signals using multi-step linear prediction based reverberation extraction.
Keisuke KinoshitaTomohiro NakataniMasato MiyoshiPublished in: ICASSP (2010)
Keyphrases
- multi step
- linear prediction
- audio signals
- spectral analysis
- image coding
- prediction error
- lossless compression
- speech signal
- sound source
- autoregressive
- signal processing
- k nearest neighbor
- subband
- speech recognition
- computer vision
- music information retrieval
- knn
- data structure
- information extraction
- filter bank
- pattern recognition
- active learning
- compression ratio
- audio signal
- semi supervised