MaryTTS Text to Speech

Description: Text To Speech plugin based on MaryTTS library

Type: - Categories: Speech Recognition & TTS

Development status: Beta version

Tested on: All platforms

Developer: Mauro Cicolella


This plugin can be used to speak any text when an event occurs. It is completely programmable using the automations system (IF this THEN say THAT).

It’s based on MaryTTS project (


Automations examples

Text to speech plugin comes with these predefined commands which can be attached to a trigger to create an automation (if you don’t know how to do read this chapter of the tutorial):

  • say current time
  • say current temperature of the thermometer
  • say electric device status
  • say door/window status
  • say readed motion level

which can be attached to any trigger to create automations like:

  • if an object is clicked -> say electric device status
  • a door becomes open -> say door/window status

...and so on

How to create new custom commands

  • Right click on an object and switch to Automations table
  • In command box write say current time or any other command starting with “say” so you can use it as a template for yours
  • Click the edit button. In the next window change the say property according to your needs (remembed you can use token substitution to insert variables like
  • Save as new command

Now your command will be available to be used in automations.

How to install better and natural sounding voices

By default the plugin uses an English female voice. You can customize it by downloading a new voice, copying the jar file into FREEDOMOTIC_ROOT/plugins/devices/marytts/lib folder and modifying the marytts-manifest.xml file as reported in the following table.

International voices
Language Voice Link Config parameters
English Female Already installed <property name=’voice-jar-file’ value=’voice-cmu-slt-hsmm-5.1.1.jar’/> <property name=’voice’ value=’cmu-slt-hsmm’/>
German Male <property name=’voice-jar-file’ value=’voice-bits3-hsmm-5.1.jar’/> <property name=’voice’ value=’bits3-hsmm’/>
Italian Female <property name=’voice-jar-file’ value=’voice-istc-lucia-hsmm-5.1.jar’/> <property name=’voice’ value=’istc-lucia-hsmm’/>

Source code

GitHub repository