multi-speaker ?

Hi @mush42 , thanks for your hard work ! Is it possible to train multiple speaker embeddings at the same time ? I'd like to experiment with ConvNeXt