Tobias Cord-Landwehr

> It seems to work well with oracle embedding (18.5dB score was improved in the WSJ-2mix validation after 50 epochs). But when two stacks are jointly trained, the separation stack...

If I remember it correctly, they also used the label shuffling in the paper. In my experiments, I did not use the architecture as proposed in the paper, but a...

I would vote against making a breaking change, here. We should keep the behavior of the function (value range between 0 and 1, but no usage of an absolute value)...

I also think that keeping the `-1` option from md-eval is not necessary. With _md-eval_, the _dscore_ python wrapper, _pyannote_ and _spyder_ there are already at least four different option...