Video Production

Audio restoration in eight easy steps using ERA-D

Fast audio repair with ERA-D

As a post-production professional and dialogue editor, you want the fastest route to audio repair, always wanting to achieve world class results. But audio repair algorithms are content depended. No matter what tool you use, sometimes you just need a different algorithm to get the job done. ERA-D is a joint denoise and dereverb plugin that has an intuitive design and state-of-the-art technology. ERA-D has patented (and therefore unique) processing algorithms. It’s the only tool that offers next-generation features such as dual microphone processing and joint denoise - dereverberation approach.

What’s Inside?

The dual microphone processing feature of ERA-D is a patented technology which enables you to repair difficult recordings previously thought to be headed for the cutting room floor. Each one of your microphones might capture different elements of acoustic scene and therefore it can provide new cues to a modern repair algorithm. You can use your boom mic to further improve the audio repair quality on your talent’s lavalier mic or use the input from your camera mic to enhance the results further. But this is just one example, you can use any secondary microphone to improve the audio repair quality of your primary source of interest.

Let's see how to use ERA-D to denoise a take from a bit of footage inside Pro Tools. In this setup, we use the additional camera mic to harvest the power of dual microphone processing.

Aux track and routing

We start by creating an Aux track. This track will be used to sum the audio coming from the actor and camera mic into a stereo signal, with each source hard-panned Left and Right respectively. We set our pan settings and route the output to the Aux track we created.

Pro Tools routing

ERA-D Signal Path and Output settings

Next, we load ERA-D as a stereo insert plug-in on this Aux track. We select

  • Denoise from the Signal Path.
  • Dual on the denoise section (to engage the dual mic processing).
  • channel L (Left) in the Output section (so that the plugin outputs only the channel source that we want to use, in this case the actress mic).
Denoise and Dual processing

Tune the denoise Knob

The playback will now output only the mic of the actress. Now we can simply turn the denoise knob clockwise to reduce noise. When "Dual" operation is on, the outer concentric ring controls noise reduction based on the cross-channel estimation of noise. We encourage you to push these knobs to extremes, starting by 3/4 of their range, without fear. They are designed to work in tandem with the other settings, which we'll optimize next.

Denoise knobs

Enable the Dereverberation module

In the real world, noise and reverb happen at the same time and interact with each other (e.g. noise “creates” reverb too). ERA-D knows this and uses reverberation reduction algorithms while denoising for better quality audio repair. The Signal Path panel allows for both Cascade and Parallel processing chains. And of course, the dual microphone mode can be enabled on the Dereverberation module as well.


Set the Range sliders

Next up, is the processing Range sliders. We start by enabling the Link button, to globally adjust the Range for all frequencies simultaneously. We tune in the right amount of processing that works best for our source. The goal here is to achieve the best possible noise reduction without overdoing the effect.

Linked thresholds

Adjust the frequency bands for better noise estimation

After finishing the global adjustment of the Range sliders, we recommend that you re-define the frequency bands. This allows ERA-D's engine to perform better noise and reverb estimations, based on the different frequency bands. The scope of this is to pinpoint noise and reverb-heavy spectral bands in order to enable more aggressive processing on these problematic areas.

Define bands

Set aggressive processing Ranges on problematic bands

With the frequency bands defined, it's time to dive in the surgical spectral band processing power of ERA-D. We disable the Link button and using the real time spectrum analyzer and the Solo toggles on each band, we apply different settings of processing Range. We get a lot more aggressive on noisy bands here, without overdoing the effect.

Solo thresholds

Tune the Intensity control parameter to further optimize the restoration

The same kind of trait applies for the Intensity control as well. Having the signal separated in meaningful frequency bands allows for proper parameter adjustment. The Intensity control is a non-linear parameter that adjusts processing intensity in a specific band. When turned clockwise, it expands suppression. So, if you hear ERA-D solidly separate noise/reverb content from dry signal on that frequency range, give the Intensity control a bump to the right.

September 13, 2017