Advice Needed - Moving to SpectraLayers Pro from iZotope RX

Hi,

I’m looking to move to SpectraLayers from iZotope RX. I can’t deal with iZotope’s abysmal support any longer.

I’ve never used Steinberg software so I’m hoping experienced users can give some honest perspective on these questions:

  1. Is SpectraLayers Pro a good alternative to iZotope RX for noise reduction? We mainly need quality voice and spectral denoising, declipping, declicking/mouth declicking, dereverb, etc. for audio post. Basically looking for a full replacement of what RX offers.

  2. What’s your opinion of Steinberg’s customer support? Can you chat with a human or is it all automated AI and “we’re too busy, you’re on your own, good luck” like iZotope?

  3. What’s your opinion of Steinberg’s licensing process? Is it a hassle with lots of hoops to jump or straightforward and quick?

Thanks for any info.

They are different beast with many commonality and features overlapping.

It’s on par with RX having a slight edge when it comes to declicking and dereverb but that’s about to change. When it comes to manual spectral repair it is clearly superior.

Robin the creator and maintainer of SLP is personally posting daily and replying to almost every post. No AI here.

It’s good, you can install onto 3 machines and run it on two simultaneously, you can deactivate and reactivate on the fly. Izotope allows you unlimited install as well as Ilok. You decide.

Check some of our videos here:
or you can watch this and try it on your own. Both have Demos

2 Likes

Hi, SpectraLayers creator here - as @Rajiv_Mudgal mentioned I’m monitoring and replying daily here.

I set up a web page with comparisons of different algorithms and different examples here :
https://divideconcept.github.io/Restoration-Comparison/
https://divideconcept.github.io/Unmix-Comparison/

3 Likes

@puffycaps I’m new to this community (3mos), but I’ve been doing NR as you describe since 2011 and have been making music in DAWs since 1998.

I have never used RX, nor much Izotope products, so I can’t speak to that. I have mainly been using Acon and Sonnox for NR…Acon for DeVerb.

I’ve known about SL for a long time, just never had a computer powerful enough to run it. Plus, what I was doing was enough for my clients. I found myself struggling to reach the level of NR I needed with those tools in the past year. SL is a complete gamechanger IMO. I should say I’m not using ARA…I’m using SL standalone for now.

And the icing on the cake is @Robin_Lobel has the willingness to interact with his user base daily on these forums. My main DAW has always been like that, where the program creator runs a forum and he will comm with the user base. Robin is different IMO, he’s willing to field harsh criticism and good at ignoring user ideas that don’t fit into his vision for SL. Thanks Robin, again, for your approach in this regard :slight_smile:

@puffycaps I suggest reading the entire SL forum section through SL10 posts to get a rundown of feelings and experiences of this SL user base and if you look into what others are looking for, you will find what folks here think could improve SL.

As far as Steinberg licencing, you need their usual suspects DL Assistant and Activation Manager…and those have their own foibles.

I use SpectraLayers Pro, which I just upgraded to version 11 from 10, to restore audio tracks. Typically from aged, damaged, or badly recorded tapes (R2R).

Addressing your questions:

1 ----------------------------------
Any signal alteration based on spectrum analysis necessitates a solid understanding of electronics and audio signal processing. Given the depth of this field of study, etc., a lot of users are looking for tools that can inject predetermined changes into the signal. This is where tools like iZotope RX originate. They are designed with the understanding that the average consumer knows very little to nothing about signal engineering. As a result, algorithms have to perform a lot of computation to figure out the user’s purpose, etc.
Another subset of users is well-versed in the audio stream itself, including its processing. Those users might search for a tool that will enable them to intercept the signal and modify it precisely as desired. This is the origin of SL. As a tool, it transforms the signal into its spectrum visual representation, allowing users to precisely operate on the visual aspects of the signal. Put another way, SL is like to Photoshop except for audio.

In conclusion, SL and iZotope RX are distinct tools that are challenging to compare with one another. SL is an outstanding spectral manipulation software that can perform all the functions of iZotope RX. This can be accomplished manually or with the use of integrated algorithms that assist in editing a spectrum.

De-noising, for instance, is a very complicated subject. This can be achieved by equalising particular ranges and thresholds in conjunction with selective gates for narrow frequencies. Static noise that obscures the valuable information might be filtered out since, regrettably, this type of de-nosing can harm a signal’s useful portion.
The fact that the noise is mingled with useful information makes this an actual issue. At this point, the spectrum subtraction algorithm becomes essential, etc. By profiling the isolated noise to feed the subtraction algorithm, this can be a very effective way to eliminate static noise from the substance of the valuable signal.
Thus far, SL and iZotop RX have both made good use of these methods. However, what happens if there is a noise fluctuating over time? I’m not aware of any efficient adaptive algorithms that really work. Not even sophisticated AI-powered systems with an abundance of highly trained algorithms can do the task. In this case, undesirable noise portions can only be masked (but not eliminated) by manual intervention into the signal. To do this, I haven’t come across a better tool than SL.
Since SL relies on a visual representation of a signal, a graphical toolkit is offered. For example, using the same technique as brush masking in Photoshop and other tools, it is simple to mask a very particular area (on a function graph, this area is frequency, time, and energy) by dimming the visible information which translates directly to the signal energy in that area.
Additionally, SL simplifies semi-3D visualisation; yet, a list of predefined colour palettes is incredibly helpful, in my opinion. Analogue noise (which comes from the transport mechanism and analogue topology), magnetic noise (occurring when a tape head begins to read magnetic domains), and any other noise (due to recording) are easily distinguished. All by simply observing the spectrum and incorporating brightness, contrast, and colour.

To sum up, SL has some built-in algorithms as iZotope RX provides, but it is more just that. SL is a very specialised tool which produces outstanding results of manual and precise signal modifications.

2 ----------------------------------
The community beyond Steinberg’s products is knowledgeable and very helpful, and I’ve found this forum to be an excellent place to ask questions. Remember that many of the people here have extensive expertise in their fields. Because of everything above, I have not needed to post directly to the support.

3 ----------------------------------
There have been a lot of complaints regarding Steinberg’s gradient of the offer (which refers to variations between a specific product versions, etc.), but I believe that it have been well tailored. Because a use case is the main focus, it is not about any particular feature. There are individuals who pursue features that, from the standpoint of the use case, they do not really need. For my signal engineering work, I also use Cubase in Elements edition, which is perfect for such projects. If I were a composer, I would most certainly choose the Artist version. If my music became profitable, I might even upgrade to the Pro.
For this reason, I found Steinberg’s licensing is suitable and appropriate. The procedure is simple and all licenses are managed from a single application. To activate your license, purchase, redeem the code, then use Licensing Manager.

1 Like

One major difference I’ve noticed between Izotope and Spectralayers is Izotope Rx has a team of developers where they compartmentalize/allocate each department to other developers whereas Spectralayers is more-or-so one lonewolf doing mostly everything. So the development cycle for RX is a lot more managed. As a matter of fact (come to think of it) Spectralayers is the only application (that I know of/can think of off the top of my head) where one developer pretty much handles most of the application.

In the gaming industry, you have lead developers but those lead developers never do optimizations, the gaming industry specifically hires developers and then allocates different tasks to different developers and that kind of teamwork helps to make things more efficient.

SawStudio

This is fantastic, I appreciate the input, everyone. In the time that your replies came in, I received an email from iZotope support saying (again) that my issue would be handled eventually and a promo email.

This is actually the issue I’m having with iZoptope, and it’s not the first time. I’ll spare the details but it shouldn’t take several days to license the software we paid for IMO.

Thanks, Robin. This is also something I really appreciate about Reaper since switching from ProTools. My main concern is getting reasonable support when funky things happen with installation, licensing, etc.

Thanks for the tips, I have begun working my way through the forum.

I am comfortable doing manual processing as needed but yes, it is nice to use the RX adaptive algorithms when they work, which is well over half the time in my experience.

This is great to hear, thanks for the info.

Good point, it does raise some concerns about what would happen to SL if Robin were to leave the company. It might not be exactly the same, but I remember interacting with the developer of SoundParticles in the early days, then something happened with the company and it’s a totally different kind of corporate vibe.

I’m going to give SL a try. Aside from noise reduction I am also very curious to play with the unmixing features.

Thanks again.

2 Likes

@puffycaps to clarify your last concern, SpectraLayers is my product and my IP - Steinberg act as my distributor here.

3 Likes

Got it, sorry for my misunderstanding and thanks for the clarification.

1 Like

Just wanted to bring to your attention that for some features like un-mixing, a pc with the right video card are near essential for fast computing.

1 Like