There is no need to restart it... Someone mentioned on the github repository that for Arch Linux and other distros the TTS engine I am using (SpeechSynthesis api) is not linked.. If you are running linux depending on your distro that is probably causing the bug