Voice HTTP Posts API

The Bulk Voice API allows programmers to send a Voice Message to one or
many cellphone or landline numbers.
However because the API requires a sound file, integrator will be required to
create an ftp account, within the LogicSMS setup and ftp the file to our servers.

http://www.logicsms.co.za/voiceAPI.aspx

This post sends a Voice message through LogicSMS
Parameters
• Username (eg. SMS@LogicSMS)
• Password
• Text1 ( Must be URL Encoded)
• VC1 (The name of a file uploaded via ftp with the extension .wav, the
number of a previously used file may also be specified)
• Text2 ( Must be URL Encoded)
• VC2 (The name of a file uploaded via ftp with the extension .wav, the
number of a previously used file may also be specified)
• Text3 ( Must be URL Encoded)
• DTMF (The number of seconds to wait, after the last clip has been played,
for a key press to be recorded by the user.
• Telnos (A comma separated list of cellular / land line numbers max 1000
characters)
• MachDetect (This parameter can be either Y /N, if set to Y the system will
wait 1-2 seconds to determine if an Answering Machine or Human has
answered the call, if its an answering machine, the system will wait for the
welcome message to finish playing and will then play the voice clip.)

Sequence

The message played to the recipient will be a continuation of 
<ANSWERING
MACHINE DETECT><Text1><VC1><Text2><VC2><Text3><WAIT FOR DTMF>

Text1, Text2, Text3 is text that will be converted to a voice and are optional
VC1 is an optional voice clip, this step does not listen for key presses.
VC2 is an optional voice clip and listens for key presses, if a key is pressed it will
interrupt the voice clip.

Example

Should you wish to send a voice clip to people in your database telling them that
they have one a prize of a specific rand value the HTTP post may look as
follows.

http://www.logicsms.co.za/VOICEAPI.aspx?username=user@org&password=de
mo&telnos=12345678&Text1=Mr
Smith&VC1=RecordYouHavewon.wav&Text2=10000
Rand&VC2=RecordThanks.wav&DTMF=3&Machdetect=Y

Return Resultset

The resulting XML looks as follows

<?xml version="1.0" encoding="utf-8" standalone="yes" ?>
<VoiceAPI><Message>OK</Message><Clips><Clip id="1">10116</Clip><Clip
id="2">10117</Clip><Clip id="3">0</Clip></Clips></VoiceAPI>

The Numbers passed back in the clip parameters are unique numbers
referencing the voice clips supplied, therefore to speed up the process during
subsequent sends the integrator should no longer refer to the clip as
filename.wav but as the unique number, for example:


http://www.logicsms.co.za/VOICEAPI.aspx?username=user@org&password=de
mo&telnos=12345678&Text1=Mr Smith&VC1=10116&Text2=10000
Rand&VC2=10117&DTMF=3&Machdetect=Y


Voice Clip Format

The voice clip should be recorded as a wav file (mp3 and wma also work).

PostBack

LogicSMS can post back details of the call to a web server of your choice. Please
email support@logicsms.net to set this up.
The parameters passed back are
• Billsec (time of call)
• Result (The result of the call)
• DTMF (List of keypresses)
• Recipient (HUMAN or MACHINE)