I posted a similar but separate question about an audio recording/mixing project in a separate post.
I’d like to record multiple people speaking separately, and also mix them into a single track for live streaming.
The use case is to record public meetings for my county. For most of the meetings, I would be able to attend in-person, but for some of them I’d like to be able to handle it entirely remotely, either by myself or a member of my staff.
For this project, I’d be allowed to leave some hardware on-site.
I’m curious if it would make more sense to use a PC with as many PCI sound cards as necessary, or if I should use dedicated hardware. I’d have to have a PC there regardless to extract the audio tracks without having to go down there to physically retrieve them.
I’d need 10+ separate input channels, each with their own track.
USB audio interfaces mostly go to eight mic inputs on the interface with eight more in an expansion. Scarlett 18i20 + Octopre tracking into whatever the DAW is, for example. Not sure what would motivate PCIe cards here or if audio interfaces are what you’re referring to by dedicated hardware.
we stream our commission meetings to youtube. we do have some dedicated audio equipment and a couple webcams but we do not have to store anything locally as youtube stores them for playback and this is within compliance for our county.
My vote for 32 bit recording so that you not being there is not ever going to result in low quality recording. Mixpre and xoom both offer products. That is the oy way to guarantee great results without having to watch your levels. That said, it’s people talking, and much of the magic of 32 bit [floating point not integer, forgot to specify] can be replaced by compressor/limiter, and some automated level control for inputs was the primitive way to make a far less favorable bet on getting a usable take. Finally, you could always do bunch of sound checks and try to idiot-proof it, but hey, it is government work so be ready to get the opposite of your intended outcome as a result - statistics are nearly guaranteeing this, which is why I wrote my recommendation for 32bit float audio, you literally can not screw it up. Whispers that don’t even show a level to smashing it past a brick wall, both will be perfectly amazing in post, as good as ideal input gain for pre clipping peaks. They made them just for this.