Combining HMM-based melody extraction and NMF-based soft masking for separating voice and accompaniment from monaural audio.
Yun WangZhijian OuPublished in: ICASSP (2011)
Keyphrases
- audio recordings
- negative matrix factorization
- multimedia
- emotion recognition
- hidden markov models
- music information retrieval
- polyphonic music
- nonnegative matrix factorization
- visual information
- content based music retrieval
- music retrieval
- signal processing
- information extraction
- image classification
- audio signal
- text to speech
- feature selection
- voice activity detection
- information retrieval