Hello,
I would like to know if an integration of IBM Watson can be done in the IVR Flows that PureCloud has.
Thank you
I'm not terribly familiar with IBM's APIs, but I believe they are REST/JSON based. If that's the case, they should work with Data Actions.
And do you know of any other integration with TTS, example, Amazon Polly or Google?
Ah, you didn't specify that you were looking for an audio integration. There isn't currently support for 3rd party IVR audio integrations. https://help.mypurecloud.com/articles/purecloud-supported-languages/
ASR and TTS are provided by proprietary technology. PureCloud does not currently support third party vendors.
Thanks for the response, another consult.
There is some way to load the audios in the Architect User Promps via API C #.
Example, create a wav audio and upload it, but through the PureCloud API.
Yes, you can record audio to a file using any 3rd party software/service you like and programmatically load them into a flow using the API, API SDK, or Architect SDK.
OK, can you tell me which API creates the Prompt and which API loads the pre-recorded audio?
Also in the Architect can you choose a Pompts automatically?
I saw that the ToAudio exists () but I can not concatenate a string, the idea is that it is dynamically.
example
audio1.wav
audio2.wav
Task.variable = 2
ToAudio (Prompt.Audio + variable)
or something similar.
OK, can you tell me which API creates the Prompt and which API loads the pre-recorded audio?
Use the architect api to create the prompt & prompt asset. A prompt asset has a uploadUri that you can use to upload the recording for it. Please notice these uris have a 1 hour limit (so if you want to upload a different recording the next day, you will need to fetch the prompt asset again to get a fresh uploadUri to use).
https://developer.mypurecloud.com/api/rest/v2/architect/index.html
Also in the Architect can you choose a Pompts automatically?
Not currently but we are adding that capability
I saw that the ToAudio exists () but I can not concatenate a string, the idea is that it is dynamically.
example
audio1.wav
audio2.wav
Task.variable = 2
ToAudio (Prompt.Audio + variable)
or something similar.
Use the Append function
Append(ToAudio(Prompt.abc), ToAudioNumber(Task.variable)).
You can also build the expression using the Audio Expression Editor and you won't have to guess at the expression.
Thanks Melissa,
But the idea of the variable is that it contains the name of the prompts, since, it will be loaded dynamically., And in the same sequence mode the error throws me.
Is there any way to dynamically do the name of the prompts?
and the other, I saw the API documentation, but it does not come out as attach a file in the uploadUri, in c #
You cannot dynamically reference a prompt. We are working on adding that capability but it's not released yet. In the meantime the workaround is to use the Switch action to play the prompt.
The error is because you need to select a prompt. Prompt. is incomplete, you need Prompt.X where X is the name of a prompt.
and the other, I saw the API documentation, but it does not come out as attach a file in the uploadUri, in c #
You need to post your recording to the uploadUri. The SDK does not have a method to use, you need to write your own code to do the POST.
Melissa, what do you mean:
You need to post your recording to the uploadUri. The SDK does not have a method to use, you need to write your own code to do the POST.
I am sure that the POST method has to be created, but what does not work out is how to upload the wav file to the Prompts.
I do not know if it happens like this:
var file = @ "D: \ file.wav";
or through a URL.
Uploading a file to a URL is a common programming task that is outside the scope of this forum. Here's one example I found with a very quick google search: https://stackoverflow.com/questions/1131425/send-a-file-via-http-post-with-c-sharp
I've created some documentation for uploading a prompt as well as a tutorial:
This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.