Reducing Binary Masking Artifacts in Blind Audio Source Separation

Stokes T, Hummersone C, Brookes T. Reducing Binary Masking Artifacts in Blind Audio Source Separation. In: AES Convention 134. Audio Engineering Society; 2013. Paper 8853. Available from: https://aes.org/publications/elibrary-page/?id=16754

@inproceedings{Stokes2013_16754,
  author = {Stokes, Toby and Hummersone, Christopher and Brookes, Tim},
  title = {{Reducing Binary Masking Artifacts in Blind Audio Source Separation}},
  booktitle = {AES Convention 134},
  note = {Paper 8853},
  year = {2013},
  month = may,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=16754}
}

TY  - CPAPER
TI  - Reducing Binary Masking Artifacts in Blind Audio Source Separation
AU  - Stokes, Toby
AU  - Hummersone, Christopher
AU  - Brookes, Tim
T2  - AES Convention 134
M1  - Paper 8853
PY  - 2013
DA  - 2013/05/06
UR  - https://aes.org/publications/elibrary-page/?id=16754
PB  - Audio Engineering Society
LA  - en
AB  - Binary masking is a common technique for separating target audio from an interferer. Its use is often justified by the high signal-to-noise ratio achieved. The mask can introduce musical noise artifacts, limiting its perceptual performance and that of techniques that use it. Three mask-processing techniques, involving adding noise or cepstral smoothing, are tested and the processed masks are compared to the ideal binary mask using the perceptual evaluation for audio source separation (PEASS) toolkit. Each processing technique's parameters are optimized before the comparison is made. Each technique is found to improve the overall perceptual score of the separation. Results show a trade-off between interferer suppression and artifact reduction.
ER  -