Blog › Comparisons

Somuno vs iZotope RX 12: a full repair suite or one-step voice cleanup?

Hongyang Jin · November 7, 2026 · 10 min read

iZotope RX 12 and Somuno both clean up voice recordings, but they are very different tools. Choose RX if you work in audio post production or want control over every repair: it has a spectral editor, dozens of separate repair modules, and plugins that run inside your DAW or video editor. Choose Somuno if you want to drop recordings or videos onto a Mac app and get clear speech back without learning a toolkit – noise, reverb and other damage handled in one pass, whole folders at a time, for a one-time US$19.99.

We haven't run listening tests of RX, so this comparison sticks to what can be checked: what each RX edition includes, what it costs, where it runs, and how each tool handles video and batches. RX 12 replaced RX 11 in April 2026, and in July 2026 Boris FX acquired iZotope. To hear the difference, try both on the same recording. The last section shows how.

Somuno and iZotope RX 12 at a glance

Where iZotope RX is the better choice

RX is a widely used repair suite in audio post production, and Somuno doesn't try to replace it. RX gives you:

• A spectral editor. You see the recording as frequencies over time and can work on a selected region instead of treating the whole file the same way.
• A full repair toolkit. iZotope counts more than 50 tools, with separate modules for voice noise, dialogue isolation, reverb, breaths, bleed, wind and rustle, plus Repair Assistant. Each has its own controls.
• Plugins inside your editor. AU, AAX, AAX AudioSuite, VST3 and ARA. iZotope lists Logic Pro, Pro Tools, Premiere Pro, DaVinci Resolve and REAPER among the supported hosts, so you fix dialogue without leaving your session.
• Real time. Voice De-noise works in real time, and Dialogue Isolate runs offline or in real time.
• More than speech. Standard and Advanced include a Music Rebalance plugin, and Advanced adds Scene Rebalance for dialogue, music and effects.
• Windows and Intel Macs. RX runs on both. Somuno needs a Mac with Apple silicon.

Somuno has none of these. It isn't a plugin, has no spectral editing or per-band controls, doesn't work in real time, and is made for speech: it removes or distorts music and other non-speech sounds. If you edit dialogue every day, RX's control is worth its price and its learning time.

Which RX 12 edition do you need?

These are the tools that matter most for voice recordings, from iZotope's edition comparison:

RX Elements covers basic voice cleanup with Voice De-noise, which iZotope says removes “anything from air conditioners to street noise”, and De-reverb. Dialogue Isolate, which iZotope says can “salvage recordings buried under noise and reverb”, starts with RX Standard, and RX Advanced adds the rest of the dialogue tools.

As of November 2026, iZotope Plus costs US$15 a month or US$150 a year and includes RX 12 Elements, and iZotope Pro costs US$25 a month or US$250 a year and includes RX 12 Standard. Both have a 7-day free trial. RX itself has no free version: RX 12 Standard runs for 10 days with export disabled, then switches to a demo mode that inserts silence at regular intervals.

Somuno costs US$19.99 once on its website (up to 3 Macs, with a 30-day money-back guarantee), or is an in-app purchase in the Mac App Store version. The free version enhances the first 60 seconds of each file. Somuno is about to launch, so see pricing.

How Somuno is different

Instead of a chain of modules, Somuno runs one AI speech-enhancement model on your Mac's GPU, and that model handles several kinds of damage at once. You don't decide which problem to fix first.

The controls are few: General Model, or Keep Laughter, which removes noise but keeps laughter; Fast or High Quality; a “48 kHz (restore high frequencies)” option for phone or 16 kHz recordings; and a Strength slider from 0 to 100 % that mixes the original back in. That's far less control than RX, and far less to learn if you clean interviews, lectures or podcasts now and then.

Missing high frequencies show the difference. In RX they're the job of Spectral Recovery, an Advanced-only module that iZotope says “helps restore frequencies above 4k that are removed when the audio is compressed for real-time streaming”. In Somuno, rebuilding them is part of what the model does on every file. We haven't compared the results. See also how to fix muffled, clipped or low-quality recordings.

Here is Somuno's output on a podcast voice under steady background noise, with default settings and both versions at the same loudness. This is Somuno's result only. We haven't processed this clip with RX.

Podcast with background noise

A podcast voice buried in steady background noise.

OriginalEnhanced

Original

Enhanced by Somuno

In the original, a haze of noise fills every gap between words, and there's nothing above about 16 kHz. After Somuno, the gaps are dark, the speech harmonics stand out, and the top band is filled in.

Batches and video files

RX's Batch Processor is in Standard and Advanced, not Elements. For video, iZotope's support documentation lists formats RX can import, MOV and M4V among them, and the manual for an earlier version says only the audio is imported. In a video project, RX is more naturally used as a plugin inside your editor.

Somuno is built around batches. Add many files or whole folders to the queue, and each result is saved next to its original, or in a folder you choose, with “-enhanced” added to the name. For MP4, MOV and M4V files, it replaces only the soundtrack and copies the picture without re-encoding. See how to clean up a whole folder of recordings at once and how to clean up the audio in a video.

iZotope, Boris FX and the new Photon app

On July 2, 2026, Boris FX announced that it had acquired iZotope from Native Instruments. iZotope is now a business unit within Boris FX, and the announcement says all iZotope products, licenses and subscriptions remain fully active. RX 12.1, released on October 15, 2026, added support for Boris FX's license activation.

In November 2026 iZotope also released iZotope Photon, a separate product that isn't part of RX. It's the closest thing the company makes to Somuno: a standalone app, also available as an ARA plugin, that runs its AI models locally to clean up noisy audio and split recordings into stems, including up to six separate voices. On a Mac it requires Apple silicon (M1 or later) and macOS 14 or later, and it opens video files such as MP4 and MOV. We haven't tested Photon; see iZotope's site for its price and trial.

Which one should you choose?

How to compare them on your own recording

1. Pick a recording with the problem you care about. Somuno's free version enhances the first 60 seconds of each file, so use a file whose first minute has the worst noise or echo, or export a one-minute WAV clip of that part.
2. The RX trial can't export, so judge RX by listening inside it. Start with Voice De-noise and De-reverb, which every edition has, then Dialogue Isolate if you're considering Standard.
3. Process the same file in Somuno with default settings, and use its Compare sheet to switch between original and enhanced.
4. Compare at about the same loudness, on headphones. A louder version always seems better.
5. Listen to the pauses: is the noise gone, or does it pump between words? Then the consonants: S, T and F should stay crisp.
6. If a result sounds over-processed, turn it down – Strength in Somuno, each module's amount in RX – and listen again.

Whatever you use, the recording matters most. Get closer to the microphone, soften the room with rugs and curtains, record a few seconds of room tone, keep lossless files when you can, and stop processing as soon as the voice sounds clear. More examples are in the before-and-after samples.

Questions