Is there any light text to speech(speech synthesis) module for Jetson Nano?

Hi everyone!
I’d like to use speech synthesis for the result of face detection and play it with Jetson Nano.
I have tried to do speech synthesis with Google TTS(text to speech) API and play the audio by pydub.
However, when audio is playing, the face detection model does not work well.
I changed speech synthesis engine to pyttsx3, but it made same problem, too.
The model works well without audio playing, so I want to find other tts(speech synthesis) module or audio play module.
Is there any light weight module for speech synthesis or audio player fit to Jetson Nano?


We have a text-to-speech example but it is only available for Xavier platform:

For Nano, you can check our TensorRT for some idea: