Spleeter
Open-source music stem separation for Python and command-line workflows.
At a glance
- Editor scoreNot yet scored
- PricingOpen source
- Best forDevelopers building simple open-source pipelines
- Paid fromNone
- Stem typesvocals, accompaniment, drums, bass, piano, other
- Facts checked19 Sep 2026
Where it wins
- Separates audio into two-, four-, or five-stem groups.
- Supports batch processing through its command-line interface.
- Offers a Python API, Docker support, and custom model training.
Where it doesn't
- Requires a self-hosted setup rather than a hosted service.
- Exports WAV audio only.
- Has no desktop app.
Our verdict on Spleeter
Spleeter is Deezer’s open-source Python library for separating music audio into independent stems. It is aimed at developers who want to add source separation to a Python or command-line workflow, rather than people seeking a hosted service or desktop application. Pretrained models handle two-stem vocals and accompaniment, four-stem vocals, drums, bass and other, or five-stem separation that also includes piano. The stated stem types also include other, and output is WAV.
Its main strength is workflow flexibility for code-based projects. Audio files can be separated through the command line, while the Python library API supports waveform- and file-based separation. Batch processing lets users process multiple files in one command. Spleeter can be installed with pip or run using Docker, and its listed integrations include Python, Docker and FFmpeg. Developers with suitable datasets can also train custom source-separation models. These options make it a fit for building and adapting a pipeline; it is not a hosted web API or a desktop app.
The trade-off is that Spleeter is self-hosted, so teams need to run it in their own environment. Its export coverage is limited to WAV, which may not suit workflows that require other output formats. Buyers looking for a ready-to-use hosted product, a desktop interface, or broader export options should consider another tool. Developers comfortable working in Python or at the command line, especially those who value an open-source library, batch processing, and the ability to train models, are the clearer fit.
Spleeter pricing
Spleeter fact sheet
| Free plan | Not verified |
|---|---|
| Paid from | None |
| Stem types | vocals, accompaniment, drums, bass, piano, other |
| Max upload size | Not verified |
| Max track length | Not verified |
| Batch processing | Yes |
| Export formats | WAV |
| Desktop app | No |
| API access | Yes |
| Deployment | Self-hosted |
| Integrations | 3 integrations: Python, Docker, FFmpeg |
| Pricing | Open source |
| Website | github.com |
| Facts checked | 19 Sep 2026 |
Spleeter integrations
Spleeter lists 3 integrations on its own site.
- Python
- Docker
- FFmpeg
Alternatives to Spleeter
- FadrA free starting point for stem separation, remix tools, MIDI detection and API workflows.7.3
- LALAL.AIFlexible stem separation with web, app, plugin, and API access.8.6
- MoisesA broad stem-separation toolkit for musicians, with bulk processing and apps across platforms.—
See all Spleeter alternatives →
Also listed in
Used Spleeter? Be the first to review it
The editor score above is our own research. What this page doesn't have yet is a reader's view — what you used Spleeter for, what worked and what didn't. No stars are seeded and no review is paid for; an editor reads every one before it appears.
Write a reviewTwo minutes · verified accounts only · read by an editor before it appears
Featured on iTechGuides
Spleeter is listed in our Stem Separation Software directory. Add the badge to your site — it links back to this page.
<a href="https://www.itechguides.com/products/spleeter/"><img src="https://www.itechguides.com/best/badge/spleeter.svg" alt="Featured on iTechGuides" width="230" height="46"></a>
Reviewed by iTechGuides Editors · Editorial team · Updated Sep 2026
Advertiser disclosure: iTechGuides is reader-supported. We may earn a commission when you click some links. It never changes a score or a verdict. How we rank.