| [ -f speech.env ] && . speech.env | |
| echo "First startup may download 2GB of speech models. Please wait." | |
| bash download_voices_tts-1.sh | |
| bash download_voices_tts-1-hd.sh $PRELOAD_MODEL | |
| python speech.py ${PRELOAD_MODEL:+--preload $PRELOAD_MODEL} $EXTRA_ARGS $@ | |