Repairing a cello part in a recording

[Nigel Keay]

A recent recording I made has a problem with intonation in one of the movements. I'd be grateful for some opinions about the feasibility of repairing it before launching into the attempt with the possible time and expense involved.

In my string quartet, 2nd movement, the group was recorded with a couple of mics in ORTF in front of the group directed towards vln1 on the left and cello on the right. Also used a second couple in the middle just behind, between and directed at the vln 2 and viola to reinforce those instruments as necessary. So, cello is recorded predominantly on track 4 but with a certain amount of bleed onto other mics.

There's a passage from bars 4-19 where the cello plays a pedal on F oscillating octaves from its lowest F. The cellist decided to play this passage with "expressive octaves" meaning that he deliberately sharpened the top F each time. Proves to have been rather misguided, but was recorded under time pressures, and the complete circumstances are a rather long story but other 3 players live in another country so not really possible to simply redo this movement (the other movements are really quite OK).

My idea is to get the cellist (or another one if necessary) to re-record the cello track, and as well to use MW Equalizer to try and eliminate the old part in the other tracks. Seems possible using maximum settings for Q and -20db focused on the two F frequencies. Retaining the integrity of the stereo image might be a problem but for me presenting the music well with good intonation is the priority.

Here is the link to listen to the passage: second movement Any thoughts?

[Phil O]

Hmm. I recorded a few measures and ran it through Melodyne. If you can believe what Melodyne is showing, the whole ensemble is a little sharp. About 446 (reference pitch) was the setting that seemed to come closest. Those octaves on the celo actually look closer than some of the other instruments through that section. But that's not my call. The important thing is that Melodyne was able to separate the instruments in polyphonic mode. I don't know if it will do what you need, but a download of the demo might be worth your time.

Phil

[Frodo]

Bravo, Phil.

[Nigel Keay]

Thanks Phil, very interesting. 446 is certainly very high. I had another possible take of that cello octave pedal passage and the upper F gave the impression of getting up to a 1/4 tone sharp, so of course I used the less extreme version. Maybe that cello part has functioned like a giant vibrato where the top of the oscillation becomes the perceived pitch, so it's thrown the whole group sharp.

I'll have a look at Melodyne - I've another couple of projects where completely wrong notes have crept in so it could be very useful if it really does work well.

[Nigel Keay]

Thanks for mentioning Melodyne, Phil, I've been having a look at it over the last few days and I'm starting to appreciate what it can do. There's seems an investment necessary in learning how it works to understand the full potential.

In the case of my quartet it was recorded on four mono tracks, however, there's a lot of bleed from the adjacent instruments, so there it's a question of finding the right strategy. The stereo file I posted already had reverb applied; I'm finding the audio quality better to go back to the raw mono tracks, so then it becomes a question of does one change the pitch across several tracks, or is it easier to change the principal part and extract the corresponding note on the other tracks. In any event it seems like an extremely useful tool for remedial work.

[Nigel Keay]

I took the plunge and bought Melodyne Editor, seems it's going to come in quite handy on a number of projects, in fact I'll have to resist the temptation of spending too much time going back to 30 year's worth of my pieces fixing them up!

[Phil O]

Glad it's working out for you. As much as I hate pitch correcting vocals, I had to do some the other day. The vocal track had a fair amount of piano bleed, but lucky for me there was no noticeable vocal bleed on the piano track. So, I was able to adjust the vocal track only, using the polyphonic algorithm to tweak the vocal without disturbing the piano bleed. Nice!

Phil

[HCMarkus]

I was able to adjust the vocal track only, using the polyphonic algorithm to tweak the vocal without disturbing the piano bleed. Nice!

Phil

Good one Phil!