hi community's friends, I like and use espeak as voice generator for my n900, I use also espeaktime and espeakcaller too, really great applications!! just "the voice" is like robot and I'd like to listen it as human voice. I changed some parameters without success. is it possible to improve voice quality to simulate a human voice?

I can not remember where I read it, but I do recall that there are several basic methods of text to speech synthesis. espeak uses one that is known to work with low ressources, but cannot produce output that sounds human - no matter how you tweak the settings. (It does, however, produce output that's easy to understand, even if it doesn't sound good.)

Other methods may sound better but might have other drawbacks such as flexibility (usage with different languages) or even patents...

As I said, can't recall the source, but when I read it it sounded reasonable to me and I stopped playing with the parameters too much.