The newest version of Ozone is here with some cutting edge new tools, along with fantastic updates to ones you already know and love. While some of these tools may seem aimed at modern production and mastering challenges, they’ve already proven useful in my own workflow when it comes to remastering old recordings. This is especially true of the brand new Master Rebalance module which has saved a few of my projects with unavailable multitracks.
In this piece you’ll learn:
- How to approach remastering an old recording
- How the new modules in Ozone can be used in remastering
Here’s the source file for our track today, “Tiritomba,” an Italian folk song, performed by Joseph Schmidt, followed by our final remaster:
Off the bat I can hear a few things I’d like to take care of in RX:
This is fairly typical on old recordings like this. A careful frequency selection at the very beginning of the file allows Spectral De-noise to make short work of this. It’s easy to go overboard here though, so exercise caution and good judgment. In this case, 6 dB of reduction is enough to clean things up nicely without any detrimental side effects.
It’s hard to say with certainty when these were introduced into the file, but there are about six of them throughout the recording, sometimes spaced very close together as shown below. The Interpolate module is a great tool for defects like this where there is a very short duration broadband signal error. By making selections that include a little bit of the signal to either side of the dropouts, it can fill them in seamlessly. If, on the other hand, we had encountered bandlimited or longer duration errors, the tools in the Spectral Repair module may have yielded better results.
There’s a very short section of repeated audio at the very end. The excellent waveform view in RX allows me to zoom in, select, and delete the repeated section with precision.
The order in which we perform these repair steps can also be very important. If you’re not familiar with it, the RX Processing Flowchart is a valuable resource.
Here’s the repaired version for comparison:
One of the big things that jumped out at me when listening initially was the vocal. As in, it just about jumped out of the speakers and slapped me in the face. While this was not abnormal for recordings from this era, in a modern context the vocal level feels a bit jarring.
The Master Rebalance module allows you to select vocals, bass, or drums and adjust the level up or down by up to 8 dB relative to the rest of the mix. You can even stack multiple instances of the component plug-in to adjust multiple elements. Here though, we just need to pull the vocal back by a few dB for a more contemporary balance.
Here’s the file with Master Rebalance applied. It’s subtle, but notice how the vocal sits just a little further back in the mix without masking the other instruments quite so much.
At this point, I’m ready to start thinking about equalization. Listening through again, I make a mental list of the four EQ-related changes I would like to make:
- Several bass resonances in the 100 Hz–125 Hz range need taming.
- There’s a slight build-up of low-mid energy.
- At the loudest passages, the vocal can still be a bit loud and harsh in the 3kHz range.
- The top-end feels slightly muffled.
Because of the way the bass resonances and vocal harshness change over time, they are perfect candidates for Dynamic EQ, allowing us to address them only when problematic. On the other hand, the fact that the low-mid build-up and muffled top end are pervasive throughout make them more suited to a traditional static EQ.
First, I use a Dynamic EQ to tame the two bass notes, which are really jumping out, and also add a small dynamic boost to the lowest note which is getting a bit lost. Then I find the vocal region, which is still feeling a bit edgy, and add a wider dynamic dip there.
Another thing that’s hard not to notice is that this recording is almost entirely mono. Again, for the era this makes sense, and any small differences between the left and right channels are probably the result of less than ideal transfers over the years. Still, who doesn’t love a little width and spaciousness?
This is another area where we want to be careful not to go overboard. There’s no panning information encoded in this signal, so we won’t be able to recover anything there, but it would be lovely to add a slight sense of ambiance to the recording. The Stereoize function in Ozone's Imager is a fantastic tool for this, and, in fact, the new Velvet-noise decorrelation mode proved to be just the ticket here.
And with that, I think we’re there. I have a Maximizer at the end of the chain to bring the level up a bit, but it does less than 0.5 dB of gain reduction a handful of times throughout the song. Due to the acoustic nature of this song, its perceived loudness is higher than its measured level would suggest, so pushing things much louder just doesn’t make a lot of sense.
While in this context, the use of Master Rebalance was on the subtle side, I’ve achieved some astonishing results with it in my work to date. I would love to share some of these examples with you, but out of respect for the copyrights held by the artists, I must abstain. Luckily, you don’t have to take my word for it. If you haven’t already, you can download a demo of Ozone Advanced and try it out for yourself.
I hope this has given you a view into the decidedly futuristic capabilities of Ozone Advanced. When the need arises it can help you breathe new life into recordings from decades gone by in new and wonderful ways.