The Sesh.FM Stem Splitter AI Tool Is Now Available for Free

When it comes to AI tools for music production, audio separation is one of the most commonly used amongst song writers, producers and audio engineers.

Sesh.FM is the first web app to offer a stable and completely free, unlimited, and browser-based suite of AI stem splitter products. No user registration is required and downloads are completely free.

For those not familiar with the company, Sesh is a browser DAW that lets multiple users collaborate in the same project at once. They're currently beta testing their service and have a waitlist for the DAW, but to keep people happy while they wait, the team has released a suite of completely free AI stem splitting tools.

Site visitors don't even have to log in to use them. Sesh lets you upload any audio file and extract stems (vocals, drums, bass lines, piano, guitars, etc) in under a minute. The Acapella Extractor can separate lead and backing vocals, while the Drum Remover isolates percussion, separating it into individual hits like kick, snare, toms, and so forth.

Visitors can download the stems or open them in Sesh (their multiplayer DAW) for further audio editing. With over a dozen companies charging a premium for this service, it's kind of shocking that Sesh has decided to give this away for free. But it also makes sense that they're growing a loyal user base with a high-value, free product.

In this article, I'll share a bit more detail about each of their stem splitters, so you can get a feel for what sets them apart. Let's start with the all purpose web app.

Getting started with the Sesh AI Stem Splitters

The Sesh AI stem splitters all have the same interface. Their key differences come down to the kinds of stems they extract. I'll cover some basics on how all three of them work and then we'll move on to specifics for each tool.

To get started, open one of the three AI tools in your browser to load any music file (MP3, WAV, M4A, AAC) of 75mb or smaller file size. For your convenience, here are the links; AI stem splitter, Drum Remover, and Acapella Extractor.

Step 1: Drag and drop your audio file into the Stem Splitter interface (no sign-up required). It should come as no surprise that high quality audio will result in better stem outputs.

Step 2: The model will process the song in under a minute. On occasion the service does time out, probably because of high server loads. When that happens, I refresh the page and upload again. In my experience, the model always works that second time around.

Step 3: Download stem or click “Open in Sesh” to import the stems into Sesh’s free browser-based DAW. Within the DAW, you can solo, mute, and adjust each instrument layer.

This tool is designed for music producers and audio engineers who need clean, individual stems for remixing or further production. The separated stems have minimal artifacts and extremely high audio quality, so you can treat them like original multitrack recordings. Even complex elements like electric guitar or layered vocals come through clearly.

By providing fully isolated tracks, Sesh gives users maximum creative control: you can mute instruments, re-balance levels, or add effects to each part independently. In short, the Stem Splitter is an incredibly powerful web tool for extracting stems and unlocking new creative possibilities in any song.

The Acapella Extractor

The Acapella Extractor acts as an AI vocal remover. Upload a song, and Sesh’s AI will extract the vocal tracks while removing most instruments. It separates multiple layers; lead vocals, background vocals, and instrumental tracks.

People often separate the vocals our to get instrumental karaoke tracks, but there are plenty of ways to use the vocal stems too. They can be sampled and remixed on new tracks or even used for studying arrangement technique.

In the example below, I've uploaded a song that uses vocal harmonies as a kind of texture. There's no foreground or lead melody. Sesh correctly knew to isolate and assign these harmonies as backing vocals.

So here's how you can do it yourself:

• Step 1: Upload your music file (MP3, WAV, M4A) into the Acapella Extractor.

• Step 2: The AI deconstructs the track in real time, isolating the lead and backing vocals from the instruments.

• Step 3: Download the isolated vocal stems in your choice of format (MP3, WAV, etc.). You can also click “Open in Sesh” to mix these vocal tracks with new instrumentation or effects.

Sesh runs entirely in the browser with no limits, you can easily separate vocals from songs of any length. Sesh even highlights use-cases such as remixes, mashups, and karaoke. The extracted acapellas are very clean – the AI effectively minimizes background noise, giving crisp vocal tracks with very little bleed from the instruments.

Compared to generic vocal removers, Sesh’s extractor produces higher-fidelity results with extra stems. In short, it makes it trivial to generate clean vocal-only and instrumental versions for remixing, sampling, or karaoke.

The Drum Stem Separator

Their Drum Remover isolates percussion and removes it from the mix. Upload a track, and the AI identifies and extracts the original drum performance. Its standout feature is selective percussion control: instead of lumping all drums into one stem, it splits out individual hits. For example, you get separate stems for kick, snare, toms, hi-hats, and cymbals. You can then mute or remove any of these as needed, or remove them all for a fully drumless mix.

• Step 1: Drag and drop a high-quality music file (MP3, WAV, etc.) into the Drum Remover tool. (Uploading high-fidelity files yields the cleanest results.)

• Step 2: The AI processes the track in seconds. It separates the drum elements and even lets you preview each drum component individually.

• Step 3: Download the drum-less version of the track or select individual drum stems to keep or discard. You can also open the result in Sesh’s DAW to program new drum parts or collaborate.

Removing drums is useful for practice, re-recording, or fixing a mix. Because Sesh preserves the rest of the music intact, you get a professional-quality instrumental version with the drums removed. This level of per-drum control is rare: most stem splitters simply remove all drums at once.

Sesh’s Drum Remover, by contrast, gives you fine-grained control over every percussion element. The result is a clean mix where the remaining instruments maintain high fidelity and minimal background noise.

Alternatives to Sesh: Lalal AI and Soundverse AI

Lalal AI and Soundverse AI are two of the most popular online stem separation services on the internet. The audio quality of their stems are more or less on par with Sesh, but there's a few differences between these three options that are worth touching on briefly.

First, Sesh does not require users to sign up in order to split audio tracks or download the stem files. This means that anyone can use the tool anonymously and with zero friction. Splitting is relatively quick, usually around 30 seconds. You can even download the full stem collection without an account.

Lalal AI lets you split stems without an account and does so very quickly, but with several limits. First, it will only allow you to preview a couple of stems at a time and you'll have to re-render the preview each time to get different instruments. For example, you can split vocals from instrumentals, but the drum or bass preview is a separate render. If you try to download any of the stem files, they'll all you to start a paid subscription.

Soundverse requires users to create an account before uploading any audio, but does allow users to download the files. When testing with a 4mb MP3 file, it took more than a minute for the server to process our request, making it the slowest service of the three. However, on a positive note, they split the track into vocals, bass, drums, piano, guitar, and "other".

Aside from differences like speed, number of separated stems, and the free download policy, Sesh also keeps a log of your prior separated tracks. I've found this to be incredibly convenient on a few occasions, when I misplaced the audio files and was able to quickly download them again.

If you're ready to give Sesh a try, check out the main AI stem separation app from Sesh or if you need a more granular breakdown, I recommend their AI drum remover and acapella extractor.

There are no affiliate codes attached to the links in this article. We encourage you to check out our blog for more music tech news and software reviews.