It sounds terrible and inhuman, but there is definitely some similarity to the target voices. It's much easier to detect the similarity when you compare the different voices to each other saying the same phrase, instead of just listening to one in isolation

Alan Rickman seems clearer than the others.

Thanks for taking a look. Agreed Alan Rickman sounds better than others. Partly because, I couldn't harvest that many sound samples with others for training. The commercial ones have their voices recorded exclusively in studios under controlled settings (besides using concatenative synthesis, so the sound bits are real recorded streams).

I have been working on a side project that attempts to clone human voices. www.voiceclonr.com presents the voices of Alan Rickman, Carl Sagan and Robin Williams. Appreciate if you could give it a try and leave feedback.

Pretty cool. Rough and fuzzy around the edges, some of the usual speech synthesis cadence issues. Any insight into the techniques you're using to do this?

This uses http://hts.sp.nitech.ac.jp (HMM-based Speech Synthesis System). Yes, it is expected to be robotic sounding since real audio streams are not used during synthesis. Some improvements can be made I believe (larger training set and some post processing) - but don't think it will change dramatically. Thanks for trying.

Nice, I had fun listening to Robo Sagan recite some of the Pale Blue Dot speech, with some fruity additions ;)

Great! Do you think there are any use cases with this ? For example, is the quality good enough to build a news/blog reader or a fun chat box ?

Alan Rickman looks the best.