The Speaker service is the most basic of the services. It is the equivalent of the mouth or voice of the application. It’s primary role is to send audio to the sound card which is most often a Text-to-Speech synthesized output.

The Brain controls the audio input start/end timing with the Listeners so if a command is sent directly to the speaker (and not through the Brain) then the Listeners are not paused and there is a risk of the application listening to itself and attempting to act on it. Future improvements will likely limit this risk.

The minimum start options for the Speaker are:

# Start the Speaker service with standard HTTP
python3 startup.py --speaker-start --use-http

# Start the Speaker service with HTTPS (SSL)
python3 startup.py --speaker-start \
  --ssl-keypath /path/to/key.pem \
  --ssl-certfile /path/to/cert.pem