search

CorentinJ / Real-Time-Voice-Cloning

Clone a voice in 5 seconds to generate arbitrary speech in real-time
Other
52.53k stars 8.78k forks source link

How to know if the training of the synthetizer is being done in parallel with my GPU? #940

Open AlexSteveChungAlvarez opened 2 years ago

AlexSteveChungAlvarez commented 2 years ago

Hello, I have been trying to train a spanish model for the synthetizer for a month, and each time I trained the models on different datasets, it took a lot of time to finish each schedule. Right now I am training on another dataset, but my monitor shows this: image It seems to me that it is not training in parallel as it should be, and it is using just about 5% of my GPU: image Is there any param I should turn on to enable training in parallel? This is the output of the training in process: image

Bebaam commented 2 years ago

1 % would indicate GPU is not used here, but the training speed sounds like gpu is used. Maybe the speed is ok for a 2060, but there are also some bugs with training speed, which have unfortunately not really solutions yet, like #700