I know its a absolute wonky workaround but you could use a second phone and enable google speech input – or an FOSS alternative: FUTO Voice Input (Local LLM Model that works pretty great. Better than google imo. Is better finding the correct words and also putting logical punctiation. – as in when should a comma or dot appear.)

Now you enable speech input on one phone and playback the voice message of the dude on the other end. Now you got all the text.