Increasing num_speakers parameter while fine-tuning Tacotron2 model ?

[Ca-ressemble-a-du-fake]

Hi,

I am following the fine-tuning tutorial from the official doc, and I added my dataset whose speaker has nothing to do with the one that was used to initially train the model. The model is Tacotron2.
My dataset has only a single speaker but overall the model will be trained with 2 different speakers. Hence should I set num_speakers parameter to 2 ?

Thanks