Updates for AT&T Natural Voices?

[PHenry1026]

Are there updates available for AT&T Natural Voices Mike and Crystal 16khz? I have Version 1.4.

If updates are available for Version 1.4, how do I get them?

[kdwhite]

None currently, although we do have many other newer voices at
http://nextup.com/TextAloud/SpeechEngine/voices.html

[PHenry1026]

It is my understanding that Version 4.2 AT&T Natural Voices Server Text to Speech (TTS) for Windows is now available.

Is Version 4.2 AT&T Natural Voices Server Text to Speech (TTS) for Windows a different product from what we use with TextAloud?

If this is an update to Version 1.4, will this update be made available to users that have Version 1.4.

[kdwhite]

It is a complicated answer. There is a new version available from AT&T. We are unsure if it is going to offer a big improvement in quality yet, we just havne't tried it. I believe there is at least one new language (Italian I think).

The thing that makes it complex and not a slam dunk to go ahead and put out the new version is version 4.2 voices aren't compatible with version 1.4 voices. This means they can't be mixed and matched, and also no free upgrades are being allowed by AT&T.

So, if for example you had Mike, Crystal, and a couple of others, then just wanted to add one of the new 4.2 version voices, you also have to upgrade all the others, and we can't give away those upgrades. So, it is just going to be very complicated to manage. After TextAloud 3.0 is out, we'll be able to spend the time to figure out how big an improvement 4.2 is, and how we might be able to manage that upgrade process.

[geneven]

I have always admired the AT&T voices, and would be willing to trash my copies of Crystal and Mike to get the new ones. I have faith that AT&T wouldn't release new ones unless they were better. I didn't see anywhere to buy them, regardless, however.

[PHenry1026]

Greetings:

RC Dictionaries now offers updates for AT&T NV English Voices that fixes all known bugs and turns AT&T NV English voices into a professional Text to Speech presentation with virtually no errors.