This sounds great. I can’t play with it at the moment. But, it is very hard to find software which handles realtime audio with a high level of precision. In my tests of a few softwares, it was common for either the audio or video to play from partway in, losing the first instant of content. Permit me to list my ideal behaviour:
triggering an audio+video file:
always plays video from the first frame
always plays audio from the first sample
preserves sync of audio and video
if the audio or video is slow to prepare, then it is expected the latency from the trigger event will increase (the audio might wait for the video to be ready so it can be played in sync from sample 0)