On the Importance of Temporal Context in Proximity Kernels: A Vocal Separation Case Study

Yela DF, Ewert S, Fitzgerald D, Sandler M. On the Importance of Temporal Context in Proximity Kernels: A Vocal Separation Case Study. In: AES Conference: 2017 AES International Conference on Semantic Audio. Audio Engineering Society; 2017. Paper 1-2. Available from: https://aes.org/publications/elibrary-page/?id=18752

@inproceedings{Yela2017_18752,
  author = {Yela, Delia Fano and Ewert, Sebastian and Fitzgerald, Derry and Sandler, Mark},
  title = {{On the Importance of Temporal Context in Proximity Kernels: A Vocal Separation Case Study}},
  booktitle = {AES Conference: 2017 AES International Conference on Semantic Audio},
  note = {Paper 1-2},
  year = {2017},
  month = jun,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=18752}
}

TY  - CPAPER
TI  - On the Importance of Temporal Context in Proximity Kernels: A Vocal Separation Case Study
AU  - Yela, Delia Fano
AU  - Ewert, Sebastian
AU  - Fitzgerald, Derry
AU  - Sandler, Mark
T2  - AES Conference: 2017 AES International Conference on Semantic Audio
M1  - Paper 1-2
PY  - 2017
DA  - 2017/06/06
UR  - https://aes.org/publications/elibrary-page/?id=18752
PB  - Audio Engineering Society
LA  - en
AB  - Musical source separation methods exploit source-specific spectral characteristics to facilitate the decomposition process. Kernel Additive Modelling (KAM) models a source applying robust statistics to time-frequency bins as specified by a source-specific kernel, a function defining similarity between bins. Kernels in existing approaches are typically defined using metrics between single time frames. In the presence of noise and other sound sources information from a single-frame, however, turns out to be unreliable and often incorrect frames are selected as similar. In this paper, we incorporate a temporal context into the kernel to provide additional information stabilizing the similarity search. Evaluated in the context of vocal separation, our simple extension led to a considerable improvement in separation quality compared to previous kernels.
ER  -