[scott]

Hi - I want to build an IVR interface to our calendaring system and it looks like we need to play back a string of recordings as a single response to the inquiry. We expect to receive an XML response to a POST from our calendaring app and then parse it to figure out which sound files need to be concatenated and then a single response played back. Such as "9:00 a.m. is available" then branch based on a variable to a new prompt.

Can this be done and how do we do it (or at least point me in the right direction to start experimenting)? I have gone through the manual and the sample applications and the basic IVR tree makes sense and navigation is working for me.

Thanks,

Scott...

[VAD_Support]

Hi Scott,

The VAD currently does not support playing a file dynamically composing the name at runtime. You need to define the file name at design time.

So, you shoud have a Conditional component, and drop a PromptPlaback component on each branch. Then, you select which prompt to play (which branch to execute) depending on a variable. This can be packaged into a user component, so you can reuse that code all the times you need.

That's the easiest way to do it in the current state of the art.

Regards,

[scott]

Thanks for the reply. I'll give that a try, but that is going to be a very large logic tree. I'm concerned that if I chain together a lot of pre-recorded prompts that it may sound choppy. Especially, when I'm handling 30-60 concurrent call sessions.

Has anyone that you know of had any success integrating with Nuance's Vocalizer? They have a nice VXML interface and that would be a great product to build an interface component for.

Thanks,

Scott...

[VAD_Support]

Yes, I understand that it is not the best solution for big callflows, but there is no option currently...

The 3CX IVR can not be easily integrated with the Nuance TTS engine. It requires low level code changes, it can not be done just changing the VXML script.

Regards,