Match pitch and feel of two speech files of same person

Cubase
#1

I recorded a video project (woman speaking on camera) and we then realized there was a section missing, some audio content we needed to insert. So we recorded the additional audio content and I’m inserting this audio into the original audio. It’s the same person, speaking, but the pitch of the original recording is just a bit above the newly recorded audio and the “feel” isn’t the same (different mics, different room, etc.). I’m wondering if Audition has a way of helping me to do some digital magic to match these two audio recordings to match them and make them sound like the whole thing is just one long take. Does anyone know how I might do this?

Before you say it, of course we could just re-shoot the entire thing but this would include re-renting the cameras, setting up the shot, paying for the person to come back in, etc. etc. Before we go that far I thought I’d ask the question. And in case you’re wondering, the video shows the person talking, then the shot goes off to show other images while the voice continues over the images. It’s in this “other images” section where we’re inserting the new audio file. After this, the scene goes back to the person speaking and she finishes saying what she’s saying.

Any help??? I’m guessing I can use some EQ adjustments to “match” the feel as best I can, lay in some noise to “feather” it, and possibly bend down the high video’s audio and bend up the new audio’s recording with pitch bending to match the pitches but I’m not sure if this is the best way about going at this. Is there an easier or more accurate way? I do have Necture2 installed. Maybe that could do it.

Thank you.

#2

You´re in the steinberg Cubase forums…

#3

I guess this was posted to multiple forums.

I would try Fabfilter Pro-Q 2’s EQ matching feature:

https://youtu.be/9TYJD_j6u1M


For room difference, you might want to try an envelope shaper or “deverb” plugin. There’s also TDR Proximity which can make things sound closer or further away.

#4

Yes, sorry. Posted in two different forums. I’m using Premier, After Effects, and Audition for the video stuff and mastering in Cubase 9 for the audio. DJW, thank you very much for your feedback. Fabfilter Pro-2 EQ Match is a little steep for me at $179 bucks (still cheaper than a reshoot), I’ll see what I can do with manual EQ matching first.

#5

Yeah, I was thinking it from the point of it being cheaper than a reshoot. I wonder how far you’d get with the demo version?

#6

You already have a matching EQ if you have Cubase pro.
Voxengo Curve EQ is one of the included plugs.

#7

Good point, I wasn’t sure if that did it or not! Is it just as good?

#8

Something like that calls for something like iZotope RX (not cheap)
Ambience vice you could try looking for a spike in the original audio, and try making a impulse response file, that you could feed a convolution reverb. Best done on purpose in every room the recording is going to happen on the remote location. Or simply deVerb any ambience on both recordings and add a reverb to give it some life back.
You could possibly audition RX, I think there is a free trial period, you can do all that and a lot more

#9

De-verb. EQ. Compression… and, if speaking pitch is different (as happens), you might need formant correction… Antares has a cheap version of AutoTune for $49 that might do the trick or you can fool with the pitch shifter in Cubase Audio Event Processing.