Hi there!
Here is what I would like to do. When the caller calls us, the call should go straight to voicemail for the caller to leave his name. After he has finished the voicemail recording I want the call to go into our queue.
Is this possible? How?
Thanks for your help
Chris
Why would you want to do that? They use case for this must be pretty specific, since most of the time, it happens the other way aroundā¦
Because I need to get the callers name without someone actually picking up the phone. The voicemail message with the recorded name will be processed by speech to text software.
In that case, you donāt want to use Voicemail. Getting the name out of the recording is going to be a huge challenge.
Iād start looking at the commercial modules. One of them (Queue_Pro, for example) might have a capability to record a name and STT process it. Talk to Sales - they know how to handle this kind of request.
BTW - you know we donāt have a STT module available in FreePBX natively, right?
Iām pretty sure you are going to end up writing a custom context for this, but you never know - one of the modules might get you close.
Thanks Dave! Once I have the voicemail as a wav or mp3 file the processing will be done with the help of c sharp 
So, is there no way to redirect the call to the queue after voicemail?
Thatās called call screening. Take a look at any extension, enable screening and see how it works.
Youāll probably need some custom code to make this work with a queue
Thank you! 
and there will probably be some additional custom code for that STT partā¦
To look at the context code for this, turn it on for an extension, and (from the console) ātail -F /var/log/asterisk/fullā to read what the system is actually doing. That should give you plenty of information about how to handle the incoming call if you need to work it on a queue. Also, there may be some ā*-customā context piece where you can add your own code. All of that will be in the āfillā log file.
There are quite a few 'āabstractionsā of STT for asterisk intercepting an inbound call, sending it to an STT engine, parsing the result then returning to your IVR or whatever.
https://zaf.github.io/asterisk-speech-recog/
That uses Google
Same concept with IBM
Not to steal this topic. Happy Cakeday, Dicko!
Thanks, I was wondering what that was, 10 years , wow!
nice one! Thanks!
I got it almost figured out. Do you guys know in which directory the actual sound files of the call screening are stored?
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.