It's a bit of a reach tbh. "jopa" (жопа) would be a crude way to say "ass". Then some schoolyard nicknames would sometimes have a suffix -lin or -in (-лин,-ин) as general creative addition to create nickname.
Unless I'm clueless about of some slang usage, or grandparent refered to some other Slavic language (not russian), "Joplin" (жоплин) doesn't mean anything directly.
When a user starts recording, websocket continuously sends audio and receives probable transcriptions. You can observe this process in logs live as well. Whenever the program detects a word match within a currently active window it moves the current position to the position of this match and then revisits previously processed words to see if there are any updates.
A drawback of this method is that if a user pronounces all the words wrong it won't move forward and a user might think the program simply lags. But basing the current word position on a transcription input length is too chaotic from my experience.
Or did you ask about the general idea of the site? If so - it's for automatic pronunciation check. There are many apps that do this with predefined words and phrases. With this site a user can check and practice any text. I think it's a good fit for detecting mistakes in pronunciation while preparing to make a public speech or to record an audio/video to put it on the internet. It's aimed primarily at non-native speakers, but a native speaker can find it useful for polishing their enunciation as well.