Jump to content

Recommended Posts

Posted
Hi Everyone - we've had a few requests from clients on this (mainly from old MSFT responsepoint clients or skype for biz) and always thought it was a great idea.  Anyone here done anything like it?  

Idea is:
  • Call comes in and goes direct to message
  • "Thank you for calling, please tell me who you are trying to reach?"
  • (caller) - "Brandon"
  • "Transferring to Brandon Last Name now"
Behind the scenes:
  • Message is played to caller
  • goes to a "waiting to record" callflow where we are recording that leg of the call. 
  • the recording is then converted to WAV
  • A directory has already been selected at setup, and the first name / last names are snagged and saved dynamically (so could also add "Sales" , "Support" , etc.) so that the list of potential matches is up to date with smartPBX directory
  • API Call is made to iSpeech 
  • They provide a method to send the API call with a WAV along with a list of names to compare.  Our main issue with iSpeech ASR is it is horrible at 8bit audio, but with a list to compare against and a simple name, we think it should be good.
  • iSpeech returns the name
  • name is matched against directory and a blind transfer is initiated to the user callflow or group.
  • 2 weeks later...
Posted
Hi Darren - I think I may have completely missed where you answered those Pivot questions!  Was it a different thread maybe?  I don't think I ever heard anything on this, but we'd love to report back once we get something working.  
Posted
We're in the middle of a SIPLabs-Valuphone ASR development including a callflow module for integration with Pivot.  We will update once completed.  We're currently using iSpeech but plan to include google speech as well.
×
×
  • Create New...