Dismiss Notice
We would like to remind you that we’re updating our login process for all 3CX forums whereby you will be able to login with the same credentials you use for the Partner or Customer Portal. Click here to read more.

Starting a Voice Recognition project with VAD

Discussion in 'CRM / Helpdesk / App Integration' started by jlabuelo, Dec 10, 2013.

Thread Status:
Not open for further replies.
  1. jlabuelo

    Apr 2, 2008
    Likes Received:
    Hi all

    We are gonna have to improve our 3CX PBX as now we need to include a new feature to allow users to interact with our company out of office hours.

    Pretty much the "answering system" of out of office hours would have to

    a) Ask for a code to be typed or sayed to the customer

    b) Recognize that code introduced (Typed or sayed) by the custmer and perform an SQL statement over a MySQL DB that returns 1 or 0.

    c) If the statement returned is 1 it must execute a php script and explain to the customer that the operation has been done (with a voice message pre-recorded).
    d) If the SQL returns 0 just let the customer know (with a voice message pre-recorded) that the operation can not be done.

    I have been taking a look to the forums and found that 3CX has included the VAD module, but I think for what I need (text recognition and SQL script) I would need also to use another external system. Any suggestions or orentations about where to start looking for?, did any of you already implemented such a system successfully?.

    Thanks a lot in advance!!!
  2. VAD_Support

    VAD_Support Active Member

    Aug 6, 2009
    Likes Received:
    Hi there,

    Please take a look at the following article which explains how to query a MySQL database from the VAD:

    In order to request the code to the customer, you need to use a "User Input" component. That component allow the customer to enter a sequence of digits via DTMF. Then you can play it back to the customer.

    The VAD doesn't support speech recognition yet, so there is no way to perform the recognition of spoken words. If you really need this feature, you may implement it in the following way:
    1) Record audio from customer using a Record component, and save it to a WAV file.
    2) Create a DLL and use the SpeechRecognitionEngine class from Microsoft Speech API to perform the speech recognition from the WAV audio file (http://msdn.microsoft.com/en-us/library/system.speech.recognition.speechrecognitionengine(v=vs.110).aspx)
    3) Call that DLL from an External Code Execution component.

    Hope the information helps.

    Kind regards.
Thread Status:
Not open for further replies.