ROS Simple TTS

This package provides easy-to-use wrappers for a handful of text-to-speech engines under a unified interface. TTS can be performed via a ROS service, publishing to a ROS topic, or by importing the python module and making calls directly.

The following TTS engines are supported:

• gtts
• coqui
• marytts (in progress)
• pyttsx3

Dependencies

For the sake of reducing dependencies, all supported engines are not included as dependencies (since you'll likely be using only one anyway). The engines can be acquired via pip:

pip install gtts
pip install tts
pip install marytts
pip install pyttsx3

Typical usage of these packages involves generating a .wav or .mp3 that can be read from the command line. To perform this, two operating system dependencies (aplay and mpg321) are also needed. Ubuntu systems already have aplay by default. mpg321 can be acquired via your package manager. Future plans involve removing these operating-system dependencies.

If using pyttsx3, you will likely need to install its dependency of libespeak (if on Ubuntu, this can be done via sudo apt install libespeak-dev).

Installation

From Source

This is a standard ros pacakge. Once the above dependencies have been resolved, clone into your catkin workspace and then run catkin_make.

As a binary

Coming soon, probably.

Usage

Quickstart

A launch file is provided that should work out-of-the-box:

roslaunch simple_tts simple_tts.launch

The tts_in topic should now be available. Publishing to this topic will lead to the node attempting to speak the text. As a simple demonstration, try running:

rostopic pub /tts_in std_msgs/String "data: 'hello world'"

The text should now be spoken.

Configuration

The parameter "engine" allows for different engines to be used.