Reducing mp3 file size for text to speech using api

I have a use case where I convert text to speech for my business using open AI api. I have set up conversion and translations using Make .com so that it happens with just one click. The files are then uploaded on an app. However the MP3 output files are very huge. The App I work with only accepts MP3 and has a limit on file size. Is there a way that I can reduce the size of the output file from Open AI but keep MP3 format? (Currently I’m compressing after downloading which is very inefficient)