Hi Guys!

I need to play a special sound during a long web service operation.

Application already contains a transient audio but I need to change it for a particular operation.

Is it possible to do it in OD?
I thought that it is possible with "Property" usage but in the "Data" item I cannot add a Property item (like with Prompt and Collect element).
For example, I see that it is possible to do that in transfer.


Thanks in advance for your help!
Vasily.

Transitional audio is your only bet during the long data/logic operation. You would need to add the property in a vxml node type which the Announce, P/C and Transfer belong. The Data node only supports logic.

Hi WilsonYu,

Hm...
Seems that I did not catch the entire idea.
Please confirm\disapprove...
Application has a "global" transitional audio property.
So, If I add an "Announce" node (with a new transition audio property) before "Data" node then I would need to add an additional "Announce" node after "Data" with changing back transition audio property to its "global" state.
Am I right?

Thanks for your patience!
Vasily

That's not sure. The transitional audio will only play in that particular transition between the Announce node and the data node.

Hi Wilson,

Sorry for follow up the topic but I still have some questions.

Could you please tell me if it is possible to resolve the following situation:
I have an application. I set up a transitional audio in "Item Properties" in AppRoot node. From that point this transitional audio will play during long operations in every node despite of it's type.
Now I want to switch it off in some cases during application execution.
For example application has some parameter which is a special flag and makes a decision: to play or not to play (transitional audio).
Is it possible at all?

Could you please point me the place in vxml specification where I can see that transitional property could be added in a vxml node type which the Announce, P/C and Transfer? I need to provide some explanations to a customer regarding transitional audio usage. And VXML could be a good proof in my explanations.

Thanks in advance!
Vasily.

See https://www.w3.org/TR/voicexml20/#dml6.1.1

the fetchaudio property.

to "disable" you could set the fetchaudio to the defetchaudio.wav in the data folder of the application.

Hi Ross,

Thanks for the information!
I see fetchaudio property in the specification. But sorry I still dot not understand why it is allowed only in 4 cases:
1. RootApp (global level).
2. Announce (local level).
3. P/C (local).
4. Transfer (local).


Well, I have no chance to add WSO operation somewhere else except Data Node. But I cannot add/change transient audio in Data node. Sounds like stalemate situation...

Ok.


Could you please briefly check my mad thoughts...?
Probably it is possible to do something like this:

1. Use Announce node and add in requestBegin method wso operation invocation? Is it possible at all to make a manual invocation of wso operation - just from my own java code?
Could this approach bring any results in context of my problem?

2. Does OD supports multythreading approach? I want to start in a background a long process (according to the ANI information from user) but in the same time I want to continue the main thread with customer interaction. At some point of main thread I would like to check the results and make future interaction according to the received/not received results from background operation.
Could this approach bring any results in context of my problem?

3. And could you please clarify your proposal "to "disable" you could set the fetchaudio to the defetchaudio.wav in the data folder of the application.". How I can do it in the main.flow editor? Which element I should use?

Thanks in advance for you help!
Vasily.

You add the fetchaudio using a properties item (transitional audio) on the form or announce preceeding the data node with the webservice operation.

This cannot be on a data node as it does not generate any VXML. THe Voice Browser need to know when doing a fetch if that fetch takes too long to start the audio. Given A, B, C, D E nodes where A and E are P/C nodes and B C D are data nodes. When A completes in the voice browser, the results are posted and "B" is fetched. Control does not return to the voice browser until node is reached and vxml generated. Node B forwards to node C which forwards to D, then E... It is the way it is.

In regards to the other items,

1) you are best to use the data nodes so that it is more obvious where things are happening in the application. Anything in the requestBegin for an announce node is executed BEFORE that VXML is generated and returned. So as I previously said the transitional audio needs to be on the Node doing the fetch (or prior node).

2) NO.

3) In the updatePeoperties you can remove the property or set the audio to a .01 sec wav file of silence (ddfetchaudio.wav).

Ross, Thanks a lot for clarifications.

Please do not kill me...
Given A, B, C, D E nodes where A and E are P/C nodes and B C D are data nodes.
If I set transitional audio in A and I know that in B I have a long wso operation then does it mean that the user will hear transitional audio (which was set in A) instead of silence while waiting for an answer during B execution?

Yes exactly. the Voice Browser will process the VXML generated by node A which would have the fetchaudio( or transitional audio), and apply those properties when fetching b.

Hi Ross,
Thanks a lot!

I will try it soon!

Vasily.

VasilyL, What was the result? Please share your experience. thanks