4 ms·

NMF works great for unsupervised audio source separation, e.g. [1]. See [2][3] for a good overview of the field. [1] http://librosa.github.io/librosa/generated

NMF works great for unsupervised audio source separation, e.g. [1]. See [2][3] for a good overview of the field. [1] http://librosa.github.io/librosa/generated/librosa.decompose.decompose.html http://librosa.github.io/librosa/generated/librosa.decompose... [2] https://hal.inria.fr/hal-01631185/document https://hal.inria.fr/hal-01631185/document [3] https://sci-hub.tw/10.1002/9781119279860.ch8 https://sci-hub.tw/10.1002/9781119279860.ch8

jamesb93 8y ago
If you're interested in this stuff a team i'm in proximity to are working on such problems. http://www.flucoma.org http://www.flucoma.org https://pure.hud.ac.uk/en/publications/stationarytransient-audio-separation-using-convolutional-autoenco https://pure.hud.ac.uk/en/publications/stationarytransient-a...