access icon free Single channel blind source separation based on probabilistic matrix factorisation

A novel single channel blind source separation method based on probabilistic matrix factorisation (PMF) is proposed. Compared to the conventional non-negative matrix factorisation (NMF) employing Euclidean distance or Kullback–Leibler divergence, PMF uses the log posterior probability as a cost function for optimising spectrum and activation matrices. Such cost function has an advantage that the hyperparameters are optimised numerically without cross-validation. In order to apply PMF to audio source separation, both Gaussian and Laplacian priors are considered. Exponential substitution for target matrices is also proposed to guarantee the non-negativity of the separated spectrogram. In source separation experiments, the proposed PMF-based approach provided significantly better performance than the conventional NMF.

Inspec keywords: blind source separation; matrix decomposition; probability

Other keywords: cost function; optimising spectrum; PMF; activation matrices; single channel blind source separation; audio source separation; log posterior probability; probabilistic matrix factorisation; nonnegative matrix factorisation; hyperparameters; NMF; separated spectrogram; Euclidean distance; Kullback-Leibler divergence

Subjects: Signal processing and detection; Algebra; Other topics in statistics; Signal processing theory; Other topics in statistics; Algebra

References

    1. 1)
    2. 2)
    3. 3)
      • 6. Garofolo, J.S., Lamel, L.F., Fisher, W.M., et al: ‘DARPA TIMIT acoustic phonetic continuous speech corpus cdrom’, 1993.
    4. 4)
    5. 5)
      • 3. Lee, D.D., Seung, H.S.: ‘Algorithms for non-negative matrix factorization’, Proc. Advances in Neural Information Processing Systems, Vancouver, Canada, December 2001, 13, pp. 556562.
    6. 6)
      • 4. Salakhutdinov, R., Mnih, A.: ‘Probabilistic matrix factorization’. Proc. Advances in Neural Information Processing Systems, Vancouver, Canada, December 2007, pp. 12571264.
    7. 7)
    8. 8)
      • 1. Raj, B., Virtanen, T., Chaudhuri, S., et al: ‘Non-negative matrix factorization based compensation of music for automatic speech recognition’. Proc. INTERSPEECH, Makuhari, Japan, September 2010, pp. 717720.
http://iet.metastore.ingenta.com/content/journals/10.1049/el.2017.2013
Loading

Related content

content/journals/10.1049/el.2017.2013
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading
Correspondence
This article has following corresponding article(s):
in brief