Semi-blind Speech-Music Separation Using Sparsity and Continuity Priors.
Hakan ErdoganEmad M. GraisPublished in: ICPR (2010)
Keyphrases
- audio signals
- speech music discrimination
- audio features
- speech recognition
- audio recordings
- high dimensional
- sound source
- speech signal
- sparse representation
- music composition
- digital audio
- endpoint detection
- computer music
- multi modal
- music score
- learned from training data
- recognition engine
- prior knowledge
- prior information
- bayesian framework
- prior model
- speech synthesis
- language model
- visual features
- gaussian mixture model
- maximum a posteriori
- polyphonic music
- parametric models
- automatic speech recognition
- noisy environments
- music information retrieval
- acoustic features
- music retrieval
- speaker recognition
- audio signal
- speaker identification
- broadcast news
- spoken language
- information retrieval