You can use some basic text to speech software on the robot and play sounds. Note: the robot's speaker is not very loud.

sound_play provides an API for playing sounds. We have provided a wrapper around it in fetch_api/src/fetch_api/sound.py

Run the demo

rosrun sound_play soundplay_node.py
rosrun applications sound_demo.py

API

Make sure you are running sound_play. In simulation, you will need to run this yourself. On the real robot, Create the sound object:

import fetch_api
sound = fetch_api.RobotSound()

To say something:

sound.say('Hello world')

To play a sound:

robot_sound.play_sound('E04.wav')

The sounds are stored as WAV files in the fetch_api/sounds directory. We have included several sounds from the Willow Garage sound library. Feel free to download the whole set and try other sounds.