Issue with a voice that I recorded in different rooms

[fannyr70859945]

Hello,

I'm working on a storytelling video at the moment about a young artist. The only issue that I have is that I recorded her voice with the same mic everytime, but I did record her in different houses, different room sizes etc... So the sound from those different spots sound quite different from each other. I'm doing a storytelling video where I'm switching from let's say house 1 for 5 seconds, then house 2 for 10 seconds, then going back to house 1 for 10 seconds etc... So having her voice sound different everytime few seconds makes the whole thing really weird.

I know I need to work with EQ or compressor for sure, I was just wondering if any of you knew of a tutorial that would talk specifically about this topic? I've been looking all over internet and have found nothing, but I am probably not using the right key words in my researches because there is so much storytelling videos, reports, short documentary those days that they must be some tutorials about this specific topic somewhere.

Not sure if you'll be able to help me or not, but I thought I'd give it a shot just in case

Thanks so much, have a good day!
Fanette

[SteveG_AudioMasters_ · Accepted solution]

fannyr70859945  wrote

I know I need to work with EQ or compressor for sure, I was just wondering if any of you knew of a tutorial that would talk specifically about this topic? I've been looking all over internet and have found nothing, but I am probably not using the right key words in my researches because there is so much storytelling videos, reports, short documentary those days that they must be some tutorials about this specific topic somewhere. 

I'm afraid that the reason you've found nothing is that there is nothing to find. It is simply not possible to EQ your way out of this - or apply any other fix, come to that - at all. When you record in a space, it's not just the voice you are recording, but the space as well. This takes the form of what we call 'early reflections' coming back from the walls and windows, and they become fully integrated with the speech you are recording. It's a bit like baking a cake; once you've baked it you can't unbake it again. Even in carefully designed vocal booths you get a degree of this, and because the human ear is so sensitive to human voices, it's always very easy to tell the difference between different spaces and the same speaker even in those; in houses it's even easier. If the speaker has been relatively close-miced it's sometimes possible to mask the background by adding another, but even then the changes in the voice show through.

The most advanced attempt at getting around any of this that I'm aware of is iZotope's RX suite. Even that though can only add ambience from another source to an existing recording - it won't remove the original ambience, because it can't. There is something they call a spectral deconstruction plugin that can reduce ambience a bit, but my experience of it is that it removes quite a bit of what you actually want as well, because it's not just ambience that's the issue - the voice change in the recording environment is the bit that's always going to stop this working properly. And this is only really intended for situations in extremis -  and it's not exactly cheap either, as you need the advanced version to get all of that, and it's $1,199.

So I'm afraid that there's no proper fix to be had at all. This has been plaguing sound recordists since the beginning of recording, and it's not going to go away. Once you've been thoroughly bitten by it, you plan in advance and take precautions.

[Kevin-Monahan]

Fannette,

How's the project now going considering Steve's advice? Please let us know.

Thanks,

Kevin