NextUp.com

I am able to successfully create audio exports using ATT and Microsoft voices, but when attempting to create an audio export using ScanSoft (aka Nuance or Real Speak voices--I have Samantha and Serena voices, which are part of my Dragon NaturallySpeaking installation), it creates an audio file with about 2 seconds of silence. In the debug file, I see the following error message: Set ControlTags : Control Tagging is not implemented in the Dragon NaturallySpeaking engine

There are no control tags in the text I am attempting to create an audio export for. How can I resolve this so I can create audio exports using ScanSoft voices?

The 'Control Tagging is not implemented' message you're seeing in the logs is info only. It is logged when TextAloud requests the voice engine to treat the text as tagged, to allow control tags to work. Dragon voices do not support tagging, so the request to treat text as tagged fails, and the message is logged as a result.

I have Dragon installed on a system here, and I see the same problem when writing audio files. As far as I can tell, the voices supplied with Dragon do not support writing audio files. You'd need to purchase voices for TextAloud at http://www.nextup.com/TextAloud/SpeechE ... oices.html.

Sorry I don't have a better answer ...

Thanks for your response, Jim. I was under the impression that the Nuance Real Speak voices on the page you linked to were exactly the same as the voices installed with Dragon. So if I purchased the Nuance Real Speak Samantha or Serena voices, wouldn't I be purchasing the exact same voices that come with Dragon that I already have installed?

I don't understand why voices that work with TextAloud with the Speak function do not work with the audio export function. Is it some kind of technical issues with the Real Speak voices in general or just with the Real Speak voices that come bundled with Dragon?

The only guess I have is that the Nuance voices bundled with Dragon don't support writing audio files. The voices we sell at NextUp will have the same names, and sound the same when speaking text aloud, but our voices definitely support tagging and writing audio files.