5. Configuration

5.1. sonos.conf

The configuration file for the Sonos module, sonos.conf, can be found in Homegear’s family configuration directory (default: /etc/homegear/families). In this file, you can configure the event server listening for packets from the Sonos speakers and Ivona TTS.

5.2. Event Server

For most installations the default configuration should work fine. By default it looks like this:

[Event Server]
id = My-Sonos-1234
deviceType = eventserver
#host =
#port = 7373
ttsProgram = homegear -e rs DeviceScripts/Sonos/IvonaTTS.php

host can be set to the ip address the event server should listen on, if it cannot be automatically determined. The default port is 7373. If necessary this can be changed as well.

5.3. TTS

ttsProgram is set to an Ivona script by default. Currently this is the only easy to use script available. This script needs two options which can be defined in the [General] section of sonos.conf:

  • ivonaTtsAccessKey: The access key (the short one) which you get after signing up for the Ivona TTS Cloud.

  • ivonaTtsSecretKey: The secret key (the long one) which you get after signing up for the Ivona TTS Cloud.


Note that it takes a while after signing up until access and secret key work.


If you develop your own user interface, ivonaTtsAccessKey and ivonaTtsSecretKey can alternatively be set using the RPC method setFamilySetting. This overwrites the settings in sonos.conf.

5.3.1. Test TTS

First determine the peer ID of the speaker you want to output TTS on. The speaker needs to be a master, i. e. it must not be paired as a slave to another speaker. Let’s assume the peer ID is 16, then to test TTS execute the following commands in Homegear’s CLI:

# For English
$hg->setValue(16, 1, "PLAY_TTS_LANGUAGE", "en");
$hg->setValue(16, 1, "PLAY_TTS", "Hello world");

# For German
$hg->setValue(16, 1, "PLAY_TTS_LANGUAGE", "de");
$hg->setValue(16, 1, "PLAY_TTS", "Hallo Welt");

There are two more TTS variables:

  • PLAY_TTS_VOLUME: This sets the volume used for TTS output. If set to -1, the currently set volume is used.

  • PLAY_TTS_UNMUTE: If set to true and the speakers are muted, they are unmuted before playing the TTS.