Audacity is one of the best free and open-source digital audio workstations (DAW) that many content creators use. It has now even become more powerful with the introduction of AI plugins. In this blog post, I uncover 4 free AI tools that work well with Audacity. These AI in Audacity work on your computer without the need for internet or costly subscriptions.
How to Install Audacity AI Plugins
To install and use AI plugins in Audacity, there are a few things you need to sort out. First, make sure that you have the latest version of Audacity installed on your computer. At the time of this writing, it’s Audacity version 3.5.1.
Next, download all the plugins from this link, extract them, and then copy and paste the files into the folder where Audacity is installed.
Finally, open your Audacity software then go to, Edit > Preferences > Modules > Enable mod-openvino.
Restart Audacity to make sure that the plugins are properly installed.
Plugin #1: Free AI Transcription
The first amazing plugin is an AI transcription that does its job really well. The best thing about this is that you can now use it to transcribe your content without paying a huge sum of money.
To use this feature, open a timeline on Audacity and import the audio file you want to transcribe. Then go to Analyze > OpenVINO Whisper Transcription.
On the OpenVINO Whisper Transcription tab, select Transcribe under Mode and choose the language of the file you will transcribe. As of this writing, the AI plugin supports several languages. After that, click Apply.
The plugin will transcribe your audio in a few seconds, depending on how long the source file is.
To export the transcription file, go to Export Other > Export Labels > Save it on your computer.
NOTE: This AI Transcription plugin can also translate audio files from one language to another.
Plugin #2: Free AI Background Noise Remover
The second AI in Audacity allows you to suppress noise in your audio recording. This tool is perfect for removing background noises such as wind and fan sounds.
To use this plugin, go to Effects > OpenVINO AI Effects > OpenVINO Noise Suppression. Then click the Apply button on the popup window.
The tool will automatically analyze and process your audio and remove unwanted noises.
OpenVINO vs Built-in Noise Reduction
I compared the result of using this plugin to Audacity’s original noise reduction feature and found that there is not much difference. The only advantage of Audacity’s built-in noise reduction feature is that you have better control over how much noise you can remove.
Overall, the OpenVINO Noise Suppression plugin is an amazing noise reduction tool that works with just a press of a button.
Plugin #3: Free AI Vocal Remover
The next plugin on our list surprised me when I tried it out. This tool enables users to split up a music track into different components. This means you can isolate the vocals to create an acapella version or remove the vocals for a karaoke version.
To use this plugin, go to Effects > OpenVINO AI Effects > OpenVINO Music Separation.
On the OpenVINO Music Separation tab, choose the Separation Mode that works best for your needs. You can split the music file into either 4 stems (Drums, Bass, Vocal, Others) or 2 stems (Instruments, Vocals). Then click Apply.
Each isolated component of your audio will have its track on your timeline. This allows you to listen to each track separately and make necessary changes. You can also download each track as separate files.
I was incredibly stunned by the great job this AI plugin did when I tried it out. Overall, this is one of the best free music separation AI tools out there.
Plugin #4: Free AI Music Generator
The final AI in Audacity I want to share can help you generate music. There are two ways to do this using the plugin. The first is by using a sample music to remix into a different audio mix and the second is by using a prompt.
Remixing from a Sample
To use this method, you’ll need to open the sample audio in Audacity. Then go to Effects > OpenVINO AI Effects > OpenVINO Music Style Remix.
On the new popup window, you’ll see a lot of settings you need to set up. First, make sure to choose the Device (either CPU or GPU) to process the remixing. For the other inputs, check the list below from Audacity.
- What to remix to? prompt: This is used to describe the genre or style for how you want your music selection to be remixed.
- Num Inference Steps: The number of UNet iterations used to remix each 5 seconds of audio. In general, the higher this is set, the higher the quality of the generated audio will be (at the cost of time taken to remix). But, you may start seeing diminished results starting at 25 iterations or so.
- Guidance Scale: A value that represents how much the AI processing will adhere to your text prompt. Recommended to set between 7 and 9.
- Seed: Used to initialize the RNG (random noise generator). Set the seed to generate consistent results, or to recreate something previously generated.
- Strength: A value between 0 and 1. This is used to control how far the remixed audio is allowed to deviate away from the original. The closer the value is to 0, the closer the generated results will be to the original audio.
Click the Apply button to start the remixing process. The remixed audio will show up as a new track on your Audacity timeline.
Remixing from a Prompt
The second method allows users to generate new music from a text prompt and a few controls. To use this feature, create a new Audacity timeline then go to, Generate > OpenVINO Music Generation.
A popup window will appear and you’ll find a few controls you need to tweak. The input you’ll add to each element will affect the resulting music that the AI will generate.
Audacity has a comprehensive description of each property. See below:
- Mode: Used to select between Simple and Advanced modes
- Normalize: The decibel level to normalize the generated audio.
- Duration: The desired duration of the generated audio snippet.
- What kind of Music? prompt: This is used to describe the type of music to generate. It can be a genre, an artist, a style, etc.
- Seed Image: This is used to select between a number of pre-defined ‘base’ ~5s audio snippets that are used to define a starting structure (tempo, etc.) for the generated audio. Each ~5s of audio that is generated will be a modified form of the ‘base’ snippet.
- Strength: A value between 0 and 1. This is used to control how far the generated audio is allowed to deviate away from the ‘base’ audio snippet (the seed image described above). The closer the value is to 0, the closer the generated results will be to the pre-defined audio snippet.
- Seed: Used to initialize the RNG (random noise generator). Set the seed to generate consistent results, or to recreate something previously generated.
- Guidance Scale: A value that represents how much the generation will adhere to your text prompt. Recommended to set between 7 and 9.
As for the Devices, again choose which one (CPU or GPU) you have on your computer.
Once you click Generate, the plugin will create an audio snippet based on the prompts you added.
Further Audacity Upgrades
Audacity has always been a powerful tool for audio editing. But these 4 AI plugins take it to a whole new level. These AI in Audacity can help you transcribe audio, remove noise, isolate vocals, and generate original music for free. Give them a try and let us know how useful they are in your projects. If you are looking to level up your audio using Audacity – why not try our audio presets? With just one click you can sound better. It’s easy!
