PIT Loss for multichannel audio for speech separation

I have a 4 channel audio generated by my model (left,right,side,mid).
I can I apply PIT loss into it
The shape of the tensors are
Speaker one : [batch,channel,time]
Speaker two: [batch,channel,time]

If I need to apply PIT, how should I apply : [batch,channel,speaker,time] ?

if I convert it to mono, or take the mean, the model is unable to learn 4 channels properly.

No file, test, or entry point is named in the report. Reproduce the stated [batch, channel, time] inputs and inspect existing PIT-loss usage for multichannel separation; the issue is complete when the supported tensor layout and a verified usage example are documented.

Written by the indexing model from the issue text.