I imagine it could pair well with Cluster-Of-Clusters Analysis (COCA) when clustering audio samples from multiple sources
Show this thread
COLA is a self-supervised pretraining method for learning general-purpose audio data representations. Developed by @aaqib_saeed et al.
Paper: https://arxiv.org/abs/2010.10915
Code: https://github.com/google-research/google-research/tree/master/cola … (research done with TF/Keras)pic.twitter.com/ATiOaUDLoU
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Could it be used for eeg data?
-
One of the reasons COLA does not exploit data augmentation is to make it easily applicable to signals for which designing meaningful augmentations is less straightforward than for e.g. images. EEG is one of them.
- Show replies
New conversation -
-
-
Thanks for sharing our work François!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Thank you for understanding what this means...If you need me, I'll be over on the end-end-end user side of the pool.
#OverMyHeadThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
How good does it recognize speech/english?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I wonder that is it possible to share the pre-train encoder model so that we don't need to train it from scratch? Thanks.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.