We’re connecting to the speech server. This may take a moment; once connected, language and model selection will be ready to use.