Need help to find some site that provides great Text-to-Speech

[crystal.maiden_19]

Hi, I need help to find some sites that provides great text-to-speech, like Ivona.com, but much better sites, since Ivona's voices cannot include any mood/emotional (like: angry, happy, sad and etc.) or situational/conditional speaking (injured and/or dying)...
Is there any idea where I could find a site like that...?
And of course, for free (for software download and/or on-line demo(s))...

 

[Varine]

You can try Natural Reader. I don't think you're appreciating how hard it is to emulate vocal inflections synthetically either; something for use commercial that can do everything you want will probably hit at least a thousand dollars (and probably won't even do most of what you want anyway. Dying, really? Do you have any idea how complex that kind of vocal system is?) At best you might be able to find something that can change the pitch and toning of the voice with manual controls, but even that would be pretty iffy for free and I serious doubt commercially usable. You would be looking at a massive vocal model to develop that kind of thing (or an extremely efficient algorithmic set to accomplish it, which if you do happen to find tell me because I'll be wasting a lot of dev time working on one otherwise), which are already on a huge scale.

 

[monoVertex]

I subscribe to Varine's post, this is a huge and and potentially very profitable domain so you're in no luck if you want something for free and I doubt there's a large range of emotions that can be rendered with today's algorithms.

I think you'd be better off finding some amateur voice actor that would help you for free or for a really small price.

 

[crystal.maiden_19]

Whoa... So, there is no way I can get it for free, thanks for your information...

 

[Varine]

Not unless you intend to put in a significant amount of effort manually changing the tonal ranges. Which would be possible, but not very efficient. I'm doing research in it right now, but it's not going to be a very effective system for what you want probably, the only reason I want it is for dynamic conversation because I'm very lonely.

 

[Accname]

We know you are lonely varine, thats why we like bullying you so much.

 

[tom_mai78101]

Stop burning people. I bet you ate a lot of spicy chips.

 

[Varine]

We know you are lonely varine, thats why we like bullying you so much.

And why one day my computer will talk to me and not to you.

 

[crystal.maiden_19]

And why one day my computer will talk to me and not to you.

Awesome, I'll wait for that...
Can you make mine to talk to me too...?

 

[tom_mai78101]

Awesome, I'll wait for that...
Can you make mine to talk to me too...?

Whoa, dude. You're too fast for the computer to speak.

 

[Varine]

Whoa, dude. You're too fast for the computer to speak.

Is this some Asian thing? Because it definitely doesn't make sense here in the regular part of the world.

Awesome, I'll wait for that...

Can you make mine to talk to me too...?

Four to six years maybe. Synthesizing speech is secondary priority to a language engine.