[jicks]

I agree, it doesn't sound "lifelike" at all to me. Compared to WaveNet[1] it's day and night.

[1] https://deepmind.com/blog/wavenet-generative-model-raw-audio...

[Vinkekatten]

Wow, that was actually impressive. A few of those were hard to distiguish from human speakers. I love how they included breathing and mouth-sounds as well, really brings it to life.

[j2kun]

The "babbling" samples here are really fun. Would make for a great "Sims" language.

[justusthane]

I actually found that pretty unsettling to listen to, and verging on creepy. I felt like I was listening to a computer trying to pretend to be human.