Playback recorded microphone sound

Hi,

URL of the experiment: Pavlovia
for an online naming study, I am trying to create a small training at the beginning, where the participants are reading a word, and then the recorded word is played back to them.

Is this possible without having to load the recording back from the server, but by simply using a copy held in memory?

If the above is not possible, how can I load a previously made sound recording? Part of my confusion stems from not really understanding when recordings are saved online (at the end of each trial, or once at the of the experiment?), and where they are saved to. Locally, a new folder is created. Online, the recordings are not visible on the git server, I can only see them once I click on download results.

To give you an idea of what I want to achieve, I have created a dummy, see URL above

Any help would be very much appreciated.