Audacity can edit audio with AI through the free OpenVINO AI Tools plugin, which adds noise suppression, stem separation, transcription, music generation and audio upscaling to the editor you already use — and runs entirely on your own computer, with no subscription, no upload and no internet connection.
That last part matters more than it sounds. Most AI audio tools today are web services: you hand over your recording, wait for a queue, and pay per minute or per month. These effects process locally, on your machine, on files that never leave it.
AI tools for Audacity
The five AI effects at a glance:
- Cleaning up a voice recording: Noise Suppression, tuned for spoken word.
- Pulling a song apart: Music Separation, into drums, bass, vocals and everything else.
- Turning speech into text: Whisper Transcription.
- Rescuing a low-quality file: Audio Super Resolution.
- Generating something new: Music Generation and Continuation.
How we choose: we describe what each tool actually does and where it falls short, and we flag the cost of everything — free and paid alike.
What “AI editing” in Audacity actually means
Audacity on its own is a conventional audio editor: you cut, fade, EQ and export. The AI features are not built in. They arrive as a separate plugin pack called OpenVINO AI Tools, published by Muse Group and free under the GNU General Public License v3.
The name comes from OpenVINO, Intel’s toolkit for running machine-learning models efficiently on ordinary hardware. That is the whole trick behind these effects: the models that normally live on a company’s servers are packaged to run on your own processor instead. Once installed, they appear in Audacity’s menus alongside the classic effects, and they behave the same way — select some audio, choose the effect, preview, apply.
The practical consequences are worth spelling out, because they are the reason to prefer this over a web tool. There is no upload, so nothing you record — a client’s interview, an unreleased track — leaves your machine. There is no minute cap and no subscription, so processing a three-hour podcast costs the same as processing three minutes: nothing. And there is no watermark on the output.
The trade-off is equally honest: processing happens at the speed of your computer rather than a server farm. A long file takes real time, and how much depends on your hardware.
What you need before you start
Three things, and the first one causes almost every “the effects aren’t showing up” problem.
Audacity 3.7.4 or a later 3.x release. The plugin will not appear in older versions, and it does not support Audacity 4 yet, so keep Audacity 3 for these effects. Check Help ▸ About Audacity and update within version 3 if you need to. Audacity itself is free.
The OpenVINO AI Tools plugin, installed through MuseHub. Installing from MuseHub is the route the Audacity team points people to, and it handles a step that catches people out when they install by hand: during setup you choose which models and effects you want. The models are the actual machine-learning files, and they are large, so you are not forced to download all of them if you only want one tool.
A note on your platform. The plugin pack ships as a stable Windows release and, at the time of writing, a beta build for macOS. If you are on a Mac, expect the rough edges that “beta” implies. Audacity itself runs on Windows, macOS and Linux.
You may see claims elsewhere about a specific Intel processor generation being required. The official Audacity pages do not state one, and OpenVINO is designed to run across CPUs, GPUs and NPUs, so performance varies by machine rather than there being a hard cut-off we can point to. If your hardware is older, the effects should still run — just more slowly.
The five AI effects
Noise Suppression
The one most people install the plugin for. It strips background noise out of a recording — room hum, air conditioning, street noise behind a voice — and it is specifically optimised for spoken word.
Find it under Effect ▸ OpenVINO AI Effects ▸ OpenVINO Noise Suppression. You get a suppression level and a sensitivity control, plus a Preview button. Use the preview every time. Noise removal of any kind is a trade between how much noise disappears and how natural the voice still sounds, and the point where it tips into a hollow, underwater quality is different for every recording.
What makes this different from Audacity’s classic Noise Reduction effect is that it needs no noise profile. The traditional effect requires you to find half a second of silence, capture what the noise sounds like, and only then clean the track. The AI version skips that entirely, which is genuinely useful on recordings that never have a clean moment of just-noise to sample.
Best for: interviews, podcasts, voiceovers, video dialogue.
Music Separation
Splits a finished stereo mix back into its parts. You can separate a track into vocals and instrumental, or go further and pull out drums, bass, vocals and everything else as individual tracks.
This is the effect that would have seemed impossible a decade ago. It is what makes a usable karaoke version out of a released song, lets you lift a drum break to study it, or rescues a vocal take when the only surviving file is the bounced mix.
Set expectations honestly, though: separation is reconstruction, not un-mixing. On a dense, heavily-compressed master you will hear artefacts — a faint watery shimmer around the vocal, cymbals bleeding into the “other” stem. On a sparse arrangement the results can be startlingly clean.
Best for: karaoke and backing tracks, remixing, sampling, transcription practice.
Whisper Transcription
Converts speech to text using Whisper, OpenAI’s speech-recognition model. It sits under Analyze ▸ OpenVINO Whisper Transcription rather than in the Effect menu, because it produces a label track rather than changing your audio.
You pick which Whisper model to use — the larger ones are more accurate and slower, the smaller ones faster and rougher — and either set the language or let it detect one. Because the transcript lands as labels along the timeline, you get something more useful than a wall of text: the words are anchored to the moments they were spoken, so you can navigate a long recording by reading it.
Best for: podcast show notes, subtitles, interview logs, finding the one sentence you need in a two-hour recording.
Audio Super Resolution
Upsamples audio to improve its fidelity and clarity. The clearest use case is old, low-bandwidth recordings — the example the Audacity team gives is 8 kHz material such as telephone calls, where the top end was simply never captured.
It cannot invent detail that was never there in any meaningful sense, and it will not turn a phone recording into a studio session. What it does do is make thin, muffled archive material noticeably easier to listen to.
Best for: archive tape, old voice memos, phone and video-call recordings.
Music Generation and Continuation
Generates music from a text prompt, or takes an existing clip and continues it, using MusicGen. Of the five, this is the one furthest from everyday editing work — it is a sketching and idea-generation tool rather than something you reach for in a session.
The continuation mode is the more practically interesting half: give it a few bars and let it extend them, which is a quick way to stretch a piece of underscore to fit a video, or to break a blank-page moment.
Best for: rough ideas, filling out an underscore, experimenting.
The five effects compared
| Effect | What it does | Where to find it | Best for |
|---|---|---|---|
| Noise Suppression | Removes background noise from speech | Effect ▸ OpenVINO AI Effects | Podcasts, interviews, voiceover |
| Music Separation | Splits a mix into drums, bass, vocals, other | Effect ▸ OpenVINO AI Effects | Karaoke, remixing, sampling |
| Whisper Transcription | Speech to text as timeline labels | Analyze ▸ OpenVINO Whisper Transcription | Show notes, subtitles, logging |
| Audio Super Resolution | Upsamples low-quality audio | Effect ▸ OpenVINO AI Effects | Archive and phone recordings |
| Music Generation | Creates or extends music from a prompt | Effect ▸ OpenVINO AI Effects | Sketching, underscore |
How to install it
The straightforward route is through MuseHub, the free app that installs and updates Audacity plugins and instruments:
- Update Audacity to 3.7.4 or a later 3.x release — not Audacity 4, which the plugin does not support yet. This is the step to get right first.
- Open MuseHub and find OpenVINO AI Tools.
- Start the install and choose the models and effects you want. Only take the ones you will use; the model files are large.
- Restart Audacity. The effects appear under
Effect ▸ OpenVINO AI Effects, and Whisper Transcription underAnalyze.
If the effects still aren’t there, the version requirement is the first thing to re-check — it accounts for most cases.
There is also a manual route: the plugin is open source, and release builds are published on the project’s GitHub repository. Close Audacity before installing by hand. The MuseHub path exists because it saves you matching build versions and model files yourself.
Limits worth knowing
Processing takes real time. These models run on your hardware. A long file on a modest laptop will not be instant, and there is no progress you can speed up by paying.
Noise Suppression is tuned for speech. On music it can do odd things to sustained instruments and cymbals. For a vinyl rip or a music recording, Audacity’s classic Noise Reduction — or a dedicated plugin — is usually the better call.
Separation quality depends on the source. Dense, loud masters separate worse than open arrangements.
macOS is still beta. Expect the occasional rough edge on Mac.
The models take disk space. This is why the installer asks which ones you want.
FAQ
Is the OpenVINO plugin free? Yes. The plugin is free and open source under the GPL v3, and Audacity itself is free. There is no subscription and no per-minute charge.
Why can’t I see the AI effects in Audacity?
Almost always the Audacity version. The plugin requires Audacity 3.7.4 or a later 3.x release and does not run in Audacity 4 yet. Check Help ▸ About Audacity, update or switch to Audacity 3, and restart. If you installed by hand rather than through MuseHub, also confirm the plugin build matches your Audacity version.
Does my audio get uploaded anywhere? No. Everything runs locally on your computer. Nothing is sent to a server, which is the main reason to use these rather than a web-based tool for confidential material.
Does it work on Mac? Yes, but the macOS build is currently a beta release, while Windows has a stable one.
Can it remove noise from music, not just speech? Noise Suppression is optimised for spoken word. It will run on music, but results are unpredictable on sustained notes and cymbals — use the classic Noise Reduction effect or a dedicated plugin instead.
Is separated audio good enough to release? For practice, karaoke and sampling, usually yes. For a commercial release, listen critically — reconstruction artefacts are most audible on busy, heavily-compressed masters.
Next steps
For effects and plugins in general, how to use effects and plugins in Audacity covers installation and both ways of applying an effect in Audacity 4 — for the other plugins you use, since these AI effects still need Audacity 3.
If you want the cleanest possible voice recording, the AI effect is only half the job — how you set levels and treat the room matters just as much. Our guide to removing background noise in Audacity covers both routes and when each one wins, and mixing your podcast covers what to do after the noise is gone.
Producing a podcast? Podcast sound effects covers what to layer on top once the voice is clean. Digitising records instead? Recording a turntable into Audacity has the vinyl-specific workflow.
Looking further afield, the best free audio editing software puts Audacity in context against the alternatives, GarageBand for Windows alternatives covers the PC question specifically, and the best AI plugins covers AI tools that work inside a DAW rather than in Audacity.
Get OpenVINO AI Tools and the rest of the Audacity plugin catalogue in one place — download MuseHub free.
