The short answer is: probably only good things! Voice training may take a few hours depending on the size of your dataset. Unlike other platforms, who train all models for the same amount of time, we optimize our results by training our voice models to a variety of epochs based on your dataset. So long as you can see your model on your model page and it is making progress, it will finish up soon. You can exit the page. We'll send you an email when it's done!
If your voice model hasn't finished training after 12 hours, it's likely an error occurred. If this is happening to you, please fill out a bug report and we will take care of the issue for you ASAP.