I build a Mac meeting recorder, and the whole pitch rests on one claim: your audio and its transcription never leave your machine.
The problem is that this is exactly the sort of thing everyone says. You cannot check it from a privacy policy.
So I published the source. The capture engine and the on-device transcription pipeline are readable, under GPL.
Honest question for this crowd: does that actually change anything for you? Or is source code the thing we all say we want and never open?
Thanks for release. I already left a comment in the forum post, but generally everything is working good, so I can recommend it for work and other purposes.
Like that it starts automatically after detects that your mic is active, so no meeting were forgotten.
@maksym_nocorny Thank you mate! You have been on this since the early builds, and half the bugs that are not in today's release are not there because you hit them first on a real call and told me. The auto detect is the part I use most too. Corder watches for an app taking the microphone, so a call you did not plan to record still gets recorded, and you decide afterwards whether to keep it.
Finally, a meeting recorder that doesn't make me look like I'm running surveillance on my own calls. Tried it in Meet and the system audio split is genuinely clean.
@sedanurhuls Yesss! That is the exact thing that made me and the people I talk to uncomfortable, and it is one of the reasons I decided to build Corder as a Free solution for anyone!
Recording at the OS level is a smart workaround, love that it sidesteps bot blocking entirely. One thing that would make it stickier for me though is adding a quick way to clip a thirty second highlight from the transcript and export it as a standalone audio or video snippet for sharing in slack or email.
@mustafasobgmyg Actually, that’s a brilliant idea! I’ll think about how to implement this naturally I really appreciate you!