Installing Rero Core

You can install Rero Core in two main ways:

  1. Using prebuilt distributables (Ubuntu & Raspbian).
  2. Flashing the prebuilt Raspberry Pi image.

This step can also be skipped if using the Reverb Robotics single board computer which already has Rero Core preinstalled, and a system service that automatically starts Rero Core on boot up. For info on setting up the board, see Setting up the Rero Board.

Using prebuilt distributables

Download the appropriate archive for your platform (Ubuntu x86_64 or Raspbian ARM) from the Rero Core distributables release page or table.

Once downloaded:

tar -xvf rero_core_<platform>.tar.gz
cd rero_core_<platform>

To start the server, run:

./bin/rero_server ./conf/config.ini

If everything is configured correctly, you should see a message similar to Server listening on 0.0.0.0:50052.

Using the prebuilt Raspberry Pi image

A preconfigured Raspberry Pi image is also provided. Flash it to a microSD card using your preferred imaging tool.

  • Default credentials: ubuntu:rero123.123.

After boot:

  1. Log in via HDMI/keyboard or SSH.
  2. Confirm that the rerocore service is running (or start ./bin/rero_server manually).
  3. Edit ./conf/config.ini as needed (see Configuration docs).

Verifying the installation

From the extracted Rero Core directory, run:

./bin/speech_recognition_cmdline ./conf/config.ini

Speak into the microphone and verify that recognized text appears in the terminal.

You can also test:

./bin/nlu_cmdline ./conf/config.ini
./bin/audio_player ./conf/config.ini
./bin/tts_cmdline ./conf/config.ini

to verify NLU, audio playback, and TTS respectively.