In my Orchestration Designer IVR application, I'm using speech grammars to accept speech input from the callers. I'm planning to make it as an external grammar. I would appreciate if any one could answer my below questions.

1) What are all the supported extensions for external grammars ?

2) Are there any restrictions using the pre-compiled grammars (.gram extension).

3) What is the preferred way of using external grammars.

4) Shall I copy the external grammar files under english/grammars folder ?

We are using Nuance 9.0 (Quantum). Any help would be greatly appreciated!!

1) This would depend on what the ASR engine supports. I have seen grxml and .gram used before. OD does not restrict the extension.

2) see 1

3) Why are you using external grammars if you are including them in the application. Typically the external grammars are external to the application. You can use %DDVAR_<variablename>% in the url to parameterize the URL. This is useful if you have multiple languages. Typically you would put the external grammar in data/<language>/grammars

4) see 3

Thank you for your quick response :-)

One final question - What tag-format Avaya support ? In our application we are supposed to use Nuance supported grammar of tag-format="swi-semantics/1.0". Does it work with AOD ?

That works as long as you make sure you are using remote dtmf processing. It is a setting when you configure an application.

That works as long as you make sure you are using remote dtmf processing. It is a setting when you configure an application. It will not work in simulation at all.

Thank you for the information. I really appreciate your help!

Thank you for the information. I really appreciate your help!

Hi Team,

We have a new requirement to integrate our application with Nuance speech . Could you please share some sample application for understanding how to call Nuance (External Grammar file) from Avaya OD and how to extract response from grammar file in prompt and collect node.

Thanks & Regards
Tarun

The way it works on OD is it sends the vxml which embeds the http link to the grammar file. The voice browser on EP or the voice platform is responsible for sending the grammar or link to Nuance and Nuance returns the recognition result back to the voice browser. The voice browser then post the result values to OD and OD store the values in the variable that OD has generated for the Prompt&Collect node.

Hi Wilson,

Thank you for your input . What I understand , just need to add voice grammar file in Prompt and collect node and EP is responsible to send the link of grammar file to nuance for execution .

I was thinking we need to call external grammar (full url of Nuance server and put the grammar file in nuance server directory).So now this part is clear .

Thanks for your support.

Tarun Jain