I'm working on an app to download recordings, and I'm using BatchDownloadJobs. However, I see that the speed at which they return results is quite slow (around 10 minutes for ~25 conversations. And they are very short conversations...).
I've also tried to use multithreading to have multiple jobs running at the same time (3, in my test). But the speed is the same as using just 1 job, or even slower!
Is this normal? An average of 2.5 conversations per minute would make unfeasible to download recordings massively...
There aren't any SLAs regarding transcoding time for recordings; the official stance is that they get done when they get done. Care will be able to determine if there's anything going on that's negatively impacting the performance of the process.