New Features and Improvements in the LALAL.AI Desktop App

Oksana -

When we first launched the desktop version of LALAL.AI in 2022, we aimed to bring our web platform's powerful capabilities to users' local machines. In our original release post, we detailed the initial features and functionality of the desktop app. Since then, we've been consistently updating and improving the application, ensuring it provides the same robust experience users have come to expect from our web tool.

This post highlights the key improvements we've implemented over time since the launch of the LALAL.AI desktop application, so that you have the most up-to-date information on the full scope of what the software offers in 2025.

1. Modernized User Interface

We've given the application a makeover with a modern design that's easier on the eyes and more intuitive to navigate, which means you'll spend less time figuring out how things work and more time actually splitting tracks into stems.

We've also added some nice touches like dark and light modes, so the app looks good whether you're working in a bright office or late at night.

2. Customizable Settings

While early versions provided access to core LALAL.AI features like batch upload, preview creation, and stem splitting, they lacked advanced customization options. Now, the desktop app offers the same level of control as the web version, allowing users to fine-tune their results.

You can now adjust three key settings directly in the desktop application:

1) Enhanced Processing. This feature helps improve the quality of stem separation, especially for complex audio tracks. With two modes — Clear Cut and Deep Extraction — you can choose how aggressively the app separates audio elements. Clear Cut provides a precise separation with minimal artifacts, while Deep Extraction goes deeper into the audio, potentially extracting more nuanced elements but with a slight increase in stem overlap.

2) De-Echo. This feature uses advanced algorithms and machine learning to remove unwanted echo and reverberation from vocals and voice recordings, resulting in cleaner, more natural sound.

3) Neural network selection. You can now choose your preferred neural network for stem splitting, optimizing results for different types of audio files.

How to Access the Settings

Follow these steps to find the settings and adjust them to your liking:

  1. Click the Split Files button in the in-app toolbar.
  2. Upload one or several tracks from your computer.
  3. Select the stem you want to extract.
  4. Click the Continue button under the stem options.

After that, you will see the settings list like this:

3. Stem Splitting History

Splitting history was added to offer users some peace of mind, ensuring that their processed files are always within reach. This feature provides a log of all the stems you’ve previously split, allowing you to keep track of your work and revisit past projects whenever needed.

When logged into your account in-app, every file you process is automatically saved to your history and displayed in the main app window. From there, you can view a list of all your previously split tracks, along with details like the file name, stem type, and date of processing. This makes it simple to reference past work or redownload stems if needed.

Free users can redownload their stems for up to two weeks after splitting, while users who have purchased one of the paid minute packs can access their splitting history and redownload stems for up to six months.

Here is what the stem-splitting history looks like in the desktop application:

4. Improved Accessibility

We believe in making LALAL.AI accessible to everyone. That's why we've implemented new accessibility features, improving usability for screen readers. Users with visual impairments can now easily access all in-app text and features.

5. Extended Perseus Support

We've expanded the capabilities of the Perseus neural network beyond its initial vocal separation functionality. Now, in addition to vocal stem extraction, you can also use Perseus to extract and remove Drums and Bass stems.

6. Output Format Selection

Now, you can choose the download format for stems. Supported audio formats include MP3, OGG, AAC, AIFF, WAV, and FLAC, while video files can remain in their original format or be converted to audio formats. Please note that this feature is only available to users with active minute packs.

Additional Improvements

We've been working hard to make the app run smoother and more reliably. Our team tracked down and fixed a bunch of technical glitches that could cause problems for users. This means fewer unexpected crashes, faster processing, and a more stable experience when you're splitting audio files.

Update to the latest version of LALAL.AI for Desktop, or download it from our site to try all the new features and improvements for yourself!

Follow LALAL.AI on Instagram, Facebook, Twitter, TikTok, Reddit and YouTube to keep up with all our updates.