A Freeswitch eventsocket transport for Vumi.

_ _

You can contact the Vumi development team in the following ways:

• via email by joining the the vumi-dev@googlegroups.com mailing list
• on irc in #vumi on the Freenode IRC network

Issues can be filed in the GitHub issue tracker. Please don't use the issue tracker for general support queries.

Voice transports may receive additional hints for how to handle outbound messages in the voice section of helper_metadata. The voice section may contain the following keys:

speech_url

The URL where the voice file to be played can be found. If this field is absent or None, a text-to-speech engine will be used to generate a suitable sound from the message content, otherwise this voice file will be played.

This can either be a string containing the URL, or a list of strings containing URLs to sound files that should be joined to form the message.

wait_for

Gather response characters until the given DTMF character is encountered. Commonly either # or *. If absent or None, an inbound message is sent as soon as a single DTMF character arrives.

barge_in

A boolean value that if True, stops the playback of the message when a DTMF character arrives. This allows the response to the input to be played immediately, rather than waiting for the first message to finish playing before hearing the response message. Defaults to False.

tries

If barge_in is True, this will set the number of times a message is played if no input is received. Defaults to 1.

time_gap

If barge_in is True and tries is greater than 1, this specifies the length of the pause (in ms) that is given before repeating the message, if no DTMF characters are received. Defaults to 3000.

Example:

"helper_metadata": {
    "voice": {
        "speech_url": [
            "http://www.example.com/voice/ab34f611cdee.ogg",
            "http://www.example.com/voice/cd43f622dcef.ogg"
        ],
        "wait_for": "#",
        "barge_in": True,
        "tries": 3,
        "time_gap": 5000,
    },
}