Note Clustering Based on 2-D Source-Filter Modeling for Underdetermined Blind Source Separation

Spiertz M, Gnann V. Note Clustering Based on 2-D Source-Filter Modeling for Underdetermined Blind Source Separation. In: AES Conference: 42nd International Conference: Semantic Audio. Audio Engineering Society; 2011. Paper 3-2. Available from: https://aes.org/publications/elibrary-page/?id=15941

@inproceedings{Spiertz2011_15941,
  author = {Spiertz, Martin and Gnann, Volker},
  title = {{Note Clustering Based on 2-D Source-Filter Modeling for Underdetermined Blind Source Separation}},
  booktitle = {AES Conference: 42nd International Conference: Semantic Audio},
  note = {Paper 3-2},
  year = {2011},
  month = jul,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=15941}
}

TY  - CPAPER
TI  - Note Clustering Based on 2-D Source-Filter Modeling for Underdetermined Blind Source Separation
AU  - Spiertz, Martin
AU  - Gnann, Volker
T2  - AES Conference: 42nd International Conference: Semantic Audio
M1  - Paper 3-2
PY  - 2011
DA  - 2011/07/06
UR  - https://aes.org/publications/elibrary-page/?id=15941
PB  - Audio Engineering Society
LA  - en
AB  - For blind source separation, the non-negative matrix factorization extracts single notes out of a mixture. These notes can be clustered to form the melodies played by a single instrument. A current approach for clustering utilizes a source filter model to describe the envelope over the first dimension of the spectrogram: the frequency-axis. The novelty of this paper is to extend this approach by a second source-filter model, characterizing the second dimension of a spectrogram: the time-axis. The latter one models the temporal evolution of the energy of one note: an instrument specific envelope is convolved with an activation vector, corresponding to tempo, rhythm, and amplitudes of single note instances. We introduce an unsupervised clustering framework for both models and a simple, yet effective combination strategy. Finally, we show the advantages of our separation algorithm compared with two other state-of-the-art separation frameworks: the separation quality is comparable, but our algorithm needs much less computational load, is independent from other BSS-algorithm as initialization, and works with a unique set of parameters for a wide range of audio data.
ER  -