Extract vocal, accompaniment and various instruments from any audio and video

Click Log In in the upper right corner of the LALAL.AI site page. After that, you will see a pop-up window where you can register or log in to an existing account via email, Google, or Facebook.

When you sign up or sign in via email, you receive a letter from LALAL.AI containing a one-time verification code and a temporary authorization link. Enter the code on the website to access your LALAL.AI account.

Note that the code expires 10 minutes after it is sent by email. If you haven't managed to use the code in that time, you can request a new code or simply click the authorization link in the email. The link will be valid for 24 hours. If you have not used the link within this period, you will need to re-enter your email address on the LALAL.AI website.

After you uploaded a file and received stem previews, as a new user you need to sign up to split the entire file and download full stems. Below the stem previews, select one of the options: Starter, Lite or Plus. Then click the Process the Entire File button below the option.

If you selected the Starter package, you will see the sign-up section. Enter your email or sign up with Google, Apple Id or Facebook.
Note that with the Starter pack you can split up to 10 minutes worth of audio/video and listen to the results, but won’t be able to download them. Get one of the paid packages to unlock the download feature.

If you selected the Lite or Plus package, you will see the payment methods. Select one of the options and you will be taken to the checkout page.

LALAL.AI packages do not have an expiration date per se. Once you’ve acquired a package, it’s always there until you’ve used all minutes available for splitting.

Say, you’ve bought the 90-minute Lite package and processed 90 minutes worth of audio/video through the service, then you will have to get a new package to split more files. If you’ve bought it and haven’t been using it for several weeks or even months, the package with all its minutes is still available like you’ve purchased it today.

Each LALAL.AI package comes with a limit on how many minutes worth of audio/video are available for splitting. The length of every fully split file is deducted from the package minute limit. You can split as many files as you want as long as their total length doesn’t exceed the minute limit of your package.

For example, when you get the Starter pack that comes with a 10-minute limit, you can split 10 minutes worth of audio or video in total. So, with this limit, you can split one 10-minute track, two 5-minute tracks, or about three 3-minute tracks.

There are several stem separation types you can choose from. One separation type is applied at a time, giving you two stems per file. Depending on the selected type, you receive a pair of vocal and instrumental stems, isolated drums + drumless stem, isolated bass + bassless stem, etc.

Minutes are deducted from the account by the following formula:

Total file length x stem separation type(s) number

Example:

File length: 5 minutes

Selected separation types: Drums, Piano and Vocal/Instrumental.

Total number of deducted minutes: 5 minutes x 3 stem separation types = 15 minutes

The total number of minutes you can spend on splitting stems is displayed at the top right corner of the LALAL.AI main page. Make sure you are logged in to your LALAL.AI account, otherwise you won’t see the minute count.

You can also see how many available minutes you’ve got left on the previews page. After the file upload, the service gives you the stem previews. Right below the previews you can see how many minutes and seconds the full file separation requires and how many you have at the moment.

Hover the cursor over the minute display at the top right corner of the LALAL.AI site, then click Upgrade in the appeared window. Secondly, you can scroll the site page to the pricing section and click on one of the paid options. Thirdly, purchase options are available on the stem previews page you see after your file was uploaded and processed. You can also get minutes for free by becoming our referral partner – join the referral program and receive extra minutes for every invited user.

Activation key is a unique code assigned to each premium user. With its help, you can activate your package on multiple devices/browsers, in the LALAL.AI desktop app, and use our API.

The key can be found in the profile on the LALAL.AI site (you have to be logged in to get it). It’s also available in your purchase email.

Once you’ve spent all minute credits of your pack, you need to purchase another pack to continue using premium features.

In order to enhance the quality of the results you get with LALAL.AI, consider the following tips:

1. Ensure your file is high-quality

   Better input leads to better output. Regardless of AI technology's sophistication, the quality of the original audio source is crucial in determining the final results.

   • Upload files with a higher bitrate (e.g., 320 kbps for MP3) to capture greater detail and improve separation quality.
   • Opt for lossless formats such as WAV or FLAC to achieve optimal results. These formats preserve audio quality, which is essential for precise stem extraction.
   • Select tracks in which different instruments or vocals are clearly defined, avoiding those with overlapping frequencies. Clear distinctions between sounds facilitate better separation.

2. Use the preview option

   Take advantage of the possibility to listen to stem previews. It allows you to evaluate how well the stems are being separated before committing to a full split.

   This way, you can identify any issues early on and upload another version of the track if needed without wasting minute credits on full processing that may not meet your expectations.

   Please note that by proceeding with the full split, you confirm your satisfaction with the splitting quality provided in the preview.

3. Utilize advanced features

   In LALAL.AI Stem Splitter, you can use the Enhanced Processing and De-Echo features to significantly refine audio quality by reducing various unwanted artifacts. Both features are accessible in the settings during the upload and preview stages.

   In LALAL.AI Voice Cleaner, you can apply the De-Echo feature and change the Noise Canceling Level setting to change how background noise is handled during processing. By adjusting these levels—whether Mild, Normal, or Aggressive — you can optimize the separation process based on your specific needs, ensuring that primary audio elements are more pronounced.

4. Change the neural network

   LALAL.AI neural networks operate on different models and have their own unique strengths. In the settings of the Stem Splitter and Voice Cleaner, you can experiment with the two latest available networks to see which one produces better results for your specific audio.

LALAL.AI’s de-echo feature uses advanced algorithms and machine learning to process audio signals, effectively eliminating echo and reverberation from songs, voice recordings, and vocal tracks.

To remove echo and reverb from your audio, open the LALAL.AI site. Within the upload section, click the settings icon and enable the De-echo vocals/voice feature in the drop-down menu.

Select either Vocal and Instrumental or Voice and Noise stem. Then upload your audio or video file in MP3, OGG, WAV, FLAC, AVI, MP4, MKV, AIFF, or AAC formats.

This feature works in conjunction with LALAL.AI's main stem separation networks (Phoenix, Orion, and Perseus), helping them more precisely identify all the sounds of the stem you selected for separation. Enhanced Processing is designed to give you greater control over the end result by allowing you to select from two modes: Clear Cut and Deep Extraction.

Clear Cut minimizes cross-bleeding between stems, resulting in a cleaner output, but may suppress finer details. In contrast, Deep Extraction captures more intricate details but increases the risk of cross-bleeding, which may lead to some overlap between stems.

In order to find Enhanced Processing and change its mode, follow these steps:.

1. Open the main LALAL.AI site page.
2. Within the upload section, click the settings icon at the top right corner.
3. In the drop-down menu, find the Enhanced Processing option. Select the mode, depending on your goal.

Note that Enhanced Processing can only be applied to the Vocal and Instrumental, Drums, Piano, Acoustic Guitar and Electric Guitar stems.

This feature offers three distinct levels of noise reduction to cater to a variety of audio clean-up needs:

• Mild. Applies no additional processing, allowing the audio to retain its original state with all its natural characteristics.
• Normal. Activates soft compression to subtly reduce background noise while preserving the integrity of the main audio.
• Aggressive. Implements a more forceful compression, significantly reducing noise for a clearer audio output, which may be necessary in environments with high background noise levels.

Follow the steps below to find and use LALAL.AI’s Noise Canceling Level setting:

1. Open the main LALAL.AI site page.
2. In the upload section, toggle the stem bar to expand all options.
3. Select the Voice and Noise stem.
4. Within the upload section, click the settings icon at the top right corner.
5. In the drop-down menu, find the Noise Canceling Level option.
6. Select one of the three levels depending on your needs.
7. Proceed to upload your audio or video file.

You can learn more about Noise Canceling Level here.

By default, the resulting files are provided in the same format as the original files that were uploaded. For instance, if you upload a file in FLAC format, the output will also be in FLAC format.

Premium users also have the option to choose the output format. Here's how it works:

1. Upload a file to LALAL.AI.
2. Wait for the file to be processed and the preview to appear.
3. Click the arrow on the Split in Full button to open the format selection menu.
4. Choose the desired output format from the drop-down list.
5. Click the Split in Full button to start processing the entire audio file.

Once the processing is complete, you will be able to download the resulting file in the selected format. If you don't select a format in the preview step, the results will be in the same format as the uploaded file.

Note: In addition to audio, LALAL.AI supports video file input in MP4, MKV, and AVI formats. When you upload a video file, you have two options:

• Skip the output format selection. In this case, the results will be in the original video format.
• Select an audio format. You can choose from MP3, OGG, AAC, AIFF, WAV, and FLAC. The results will be in the selected audio format.