The new TTS feature

[bnz]

Just out of curiosity: are you using the commercial Neospeech web service for this offering? The Quizlet voices are obviously the Voiceware/Neospeech voices (which are absolutely fantastic by the way). I'm asking because last time I checked they only had English available for their web service. I'm really interested in that for my own little private project. I really hope that iOS 5 gets a TTS engine on the OS side like Android has where you buy a voice once for the device and then use it across all apps. The Lion feature set luckily suggests that something might actually happen there if that is any indicator.

[Ernie]

Yes, I'm using Neospeech. And yes, their web service only offers English voices. But, I'm running my own server with their software, and I'm not really sure what I'm up against as far as keeping the server up and handling whatever traffic I get, so we'll see.

I got tired of waiting for some easy TTS solution (e.g. built in to iOS). Yeah, it should be good when it finally gets added to iOS, though it might make my own TTS work a little less worthwhile at that point. Ha.

[bnz]

Heh, anyway, great effort! I'm absolutely sure that a lot of people will like it to death. The Neospeech voices are too good not to be used for language learning.

[bnz]

www.engadget.com/media/2011/06/stevejob ... te0841.jpg

not part of the keynote, but on this screen it says "option to speak text selection". maybe something is really happening here...

[Ernie]

Your link just brought up a white page for me. It sounds like to me that the new API functions will be released today on the development site. If so, I'll see if I notice anything.

[Granzyme]

Hi!
The new TTS feature indeed is great! Is it planned to change the issue that voices can't be set back to none? I accidentally chose a wrong voice for my set and it's quite annoying...
Thanx, I totally love your App btw!

[Ernie]

Yeah, I've added the ability to set it back to "None" on my development version. If you need to reset it, you can probably do it by exporting, opening up the text file and removing the voice header, and re-importing back as a new deck.