SpectraLayers 12 – Feedback and Feature Suggestions

SpectraLayers 12 is overall a solid upgrade with some meaningful improvements. While I appreciate the new features, a few aspects left me slightly disappointed or felt underwhelming in practice.

Voice Enhancer

The Voice Enhancer tool can be useful, but results are inconsistent. In some cases, it noticeably improves clarity, while in others, it introduces artifacts or worsens the quality of the original recording.

Soundtrack Separation

The Soundtrack feature felt underwhelming in my testing. I was hoping it would perform better in isolating non-verbal vocal elements such as breaths, screams, or vocal fx — which are quite common in post-production workflows. Unfortunately, it often missed these, especially when the vocal wasn’t clearly speech. In contrast, the standard Unmix Vocals feature actually performed better in this regard.

ARA Workflow Limitations

The current ARA integration limits the potential of SpectraLayers for me, particularly in post workflows. For instance, when denoising a clip via ARA, waveform changes aren’t visible unless the edits are committed. This breaks non-destructive editing flow — there’s no easy way to undo or revert back to the original once committed (unless you use workarounds). Compared to the Direct Offline Processing (DOP) in Cubase/Nuendo, ARA lacks flexibility. With DOP, I can always remove or reorder effects and instantly return to the untouched clip.

Additionally, projects with many ARA-enabled clips tend to become unresponsive or laggy, which affects usability in complex sessions.


Suggestions for Future Improvements

  1. Brand New Special Integration for Cubase/Nuendo
    new integration method that behaves more like offline rendering, reflecting changes directly in the clip (e.g. waveform updates, cut alignment) while maintaining a connection with the SpectraLayers edits. This would significantly improve post and dialogue workflows for Cubase/Nuendo users.
    Keep the current ARA implementation for compatibility with other DAWs, but consider developing a new modern API that can be adopted across platforms.
  2. VST Versions of Common Processes
    Provide VST plugin versions of frequently used single-layer processes such as Denoise, Debleed, and similar — tools that affect only one track and do not require new layers. This would allow faster integration and better performance when working inside a DAW, similar to the approach used by iZotope RX.
2 Likes

These are all great suggestions!

For me, the following would be my top suggestions for SL13…

  1. Improved Soundtrack Separation.
  2. Improved Unmix Multiple Voices, with an auto detect voice option.
  3. Improved Unmix Song, with added separation for Background Vocals, Strings, and Synths.
  4. Add a new module called Voice Unmix where you could separate out breathing, mouth clicks, and sibilance into their own layers from the voice.
2 Likes

For me honestly my top top wish regarding future release is custom more streamline integeration other than ARA
As actually ARA in large projects like post production is a downgrade compared to DOP in term of non destructive editing as in some point using ARA in post you will bounce the track, and large projects especially in post production, lots of tracks using ARA actively slows down the application, in my opinion ARA is made primarily for music in mind usually not very long projects not extremly high number of clips/tracks

3 Likes