Genesys Cloud - Main

 View Only

Sign Up

  Thread closed by the administrator, not accepting new replies.
  • 1.  Multiple Language Support in Same Menu / Audio Sequence

    Posted 09-06-2019 16:38
    No replies, thread closed.
    Hi All, 

    I have a customer that's wanting to migrate to an all TTS IVR using Polly. We currently use  wav files to advise them of their options, and several of these are multi-language prompts for non-English speakers:  "for Spanish, press 5, for Mandarin, press 6" etc... Except in that language. 

    I'd like to do this with TTS, but not seeing it... seems like there should be a way to play one text block in English, then another in Spanish, and so on,  but it looks to me like it's locked down to just one at a time. 

    Am I missing something?
    #Routing(ACD/IVR)

    ------------------------------
    Dean Thames
    Koch Business Solutions
    ------------------------------


  • 2.  RE: Multiple Language Support in Same Menu / Audio Sequence

    Posted 09-09-2019 09:57
    No replies, thread closed.
    The only way to do that is use a bunch of tasks.
    Task 1: Play Audio for the English part, Set Language to Spanish, then jump to 
    Task 2: Play Audio for Spanish part, Set Language to Mandarin, then jump to
    Task 3: Play Audio for Mandarin, . . .
    At the end you probably want to switch back to English, then jump to a task with a Collect Input action.
    The cons of doing this (other than being complicated) is that if the caller presses a digit while the flow is still going through the list of actions, the Collect Input action will not hear it b/c it hasn't started listening yet.

    Honestly using prompts is a much better solution.



    ------------------------------
    Melissa Bailey
    Genesys - Employees
    ------------------------------



  • 3.  RE: Multiple Language Support in Same Menu / Audio Sequence

    Posted 09-09-2019 09:58
    No replies, thread closed.
    Unfortunately, it is locked down to one language at a time, and way cumbersome to flip from one flow to another in rapid succession to flip languages for each TTS prompt. In the past I have just typed my best approximation of the Spanish or French or whatever and let the English TTS slaughter it.


    ------------------------------
    George Ganahl GCP (PureCloud) ICCE CCXP
    Principal Technology Consultant
    Genesys
    ------------------------------



  • 4.  RE: Multiple Language Support in Same Menu / Audio Sequence

    Posted 09-09-2019 09:59
    No replies, thread closed.
    (I was typing and researching while Melissa was creating her answer and sent almost simultaneously...so she laid the multiple flows part out well.)

    ------------------------------
    George Ganahl GCP (PureCloud) ICCE CCXP
    Principal Technology Consultant
    Genesys
    ------------------------------



  • 5.  RE: Multiple Language Support in Same Menu / Audio Sequence

    Posted 09-09-2019 11:08
    No replies, thread closed.
    Thanks Melissa and George. 

    I'll see what I can sort out. Melissa, using prompts is problematic for us and we are trying to move away from that. I'm personally a big fan of the Polly integration as it allows us to seamlessly use Lex. Which, now that I type that, might be an option here.

    ------------------------------
    Dean Thames
    Koch Business Solutions
    ------------------------------



  • 6.  RE: Multiple Language Support in Same Menu / Audio Sequence

    Posted 09-09-2019 11:24
    No replies, thread closed.
    Looks like you can switch betwixt languages using Polly + ssml: 

    https://developer.amazon.com/docs/custom-skills/speech-synthesis-markup-language-ssml-reference.html#voice

    I'll dig into this a bit and see what happens.

    ------------------------------
    Dean Thames
    Koch Business Solutions
    ------------------------------



  • 7.  RE: Multiple Language Support in Same Menu / Audio Sequence

    Posted 09-10-2019 10:26
    No replies, thread closed.
    Ok... so it's a combo solution. Thought occurred to me that you can create .wav files of various polly voices and cobble them together to create a single prompt with multiple languages that sounds like TTS, but is actually an audio prompt. 

    There's a python sample out there that explains how to use boto3 to create an audio file that gets dropped into s3 here: 

    https://docs.aws.amazon.com/polly/latest/dg/StartSpeechSynthesisTaskSamplePython.html

    ------------------------------
    Dean Thames
    Koch Business Solutions
    ------------------------------