Samples from Mimic2 based on google's tacotron. See the code at the github repo.

This model was trained on 16 hours of an english male speaker. These samples were non training samples.

"There's a way to measure the acute emotional intelligene that has never gone out of style."

"It took me quite a long time to develop a voice, and now that I have it I'm not going to remain silent."

"The human voice is the most perfect instrument of all."

"I'm sorry Dave, I'm afraid I can't do that."