Python asr_server.py (test_ffmpeg.py) - using WEBM audio files

First time creating an issue (ever) so i'm sorry in advance for eventual mistakes or misconceptions.

After extensively testing the vosk STT capabilities for my specific context (very short utterances in several languages) it came the time to integrate it in an existent brower-based application to progressively add to it STT features.
- The idea is to have a echo server client solution where a audio file is sent to the server (by a websocket client), there the stt is done (among other things) and an "anwser" (+ a json file) is sent back to the client (web application).

If the audio acquisition (micro) was "made is python" (like i did is the testing stage) it was fine, but because the application is based on Chrome browser, the audio acquisition part already coded (javascript buttons) creates this blob file that leads to a WEBA/WEBM file (i have no control on the sample rate either).
- So, working with this paradigm, **the client needs to send the complete file (weba file) to the socket server and Then the server converts it to wav, run the vosk and then send the final/full transcription back to the client** (apart from the first chunk in a streaming approach the nexts chunks of a weba file are useless).

So my question is(are) (now i notice i'm making you guys my rubber duck): ?
- Q1- **how would my client and server sockets look like** (making the ffmpeg conversion on the server side)?
- Q2 - doesn't this case makes the use of an asinc socket based solution the wrong approach?

Notes:
- I've tested your test_ffmpeg.py client with the asr_server.py from this vosk_server repository and it works fine with weba/webm files.
- I know i am not using this for what it was best intended but for now these are the conditions i have to work with.
- The results provided by this approach (not messing with the application interface front end) can and will be used to justify redoing the front end application.

Thank you