Instant stem extraction for remixing - AI Music & Audio tool
Spleeter (Deezer) is an open-source audio source-separation toolkit that splits tracks into 2, 4, or 5 stems via pretrained TensorFlow models. It's ideal for producers, DJs, and researchers who need local, scriptable stem extraction without vendor lock-in. Deezer distributes model weights, a CLI, Python API and a Docker image; there are no paid plans from Deezer - the project is free under an MIT-style license.
Spleeter (Deezer) is an open-source audio source separation tool that extracts individual stems (vocals, drums, bass, accompaniment) from mixed audio. It uses pretrained TensorFlow models provided by Deezer Research to deliver reproducible separation for music production, remixing, and music information retrieval tasks. The key differentiator is a lightweight, scriptable workflow: a command-line interface, Python API and an official Docker image that enable batch processing and integration into DAW tooling. Spleeter is freely available under a permissive license, making it accessible to hobbyists and professionals in the AI Music & Audio space.
Spleeter is an open-source source-separation toolkit published by Deezer Research in 2019. Designed as a research-to-practice release, Spleeter provides pretrained TensorFlow models and tools to split audio into discrete stems for downstream audio work. Deezer positioned Spleeter to accelerate research in music information retrieval while giving creators a practical local toolset.
Because the project is available under a permissive open-source license and distributed on GitHub, users can audit models, run them locally, and incorporate the models into batch audio pipelines without vendor-hosted subscriptions. Spleeter ships with several concrete features. It includes pretrained TensorFlow models for 2-, 4- and 5-stem separation (vocals, drums, bass, and accompaniment variants).
A command-line interface allows commands like: spleeter separate -i input.mp3 -p spleeter:4stems -o output_dir to process single files or entire directories. The Python API exposes Separator objects and save_to_file methods for scripted workflows and integration into notebooks or DAWs. Deezer also publishes an official Docker image (deezer/spleeter) so teams can containerize separation on servers or CI.
The tool accepts common formats (MP3, WAV) and supports batch processing via CLI or scripted loops. Pricing-wise, Deezer provides Spleeter as a free, open-source project with no official paid tiers. There is no hosted subscription from Deezer; the software and pretrained models are available at no charge under an MIT-style license.
Organizations that need managed, scalable hosting must either self-host Spleeter on their cloud infrastructure or purchase hosted separation services from third-party vendors who may wrap Spleeter in a paid product. Enterprise costs therefore depend on self-hosting infrastructure or third-party vendor pricing - Deezer itself does not sell a pro plan. Spleeter is used by remixers, DJs, audio engineers, music technologists, and MIR researchers.
Example workflows include: a DJ/Remix Producer using Spleeter to extract vocal stems for a new bootleg remix, and a Music Researcher using batch separation to generate isolated instrument corpora for training models. Podcast editors use it to remove musical beds, while archivists separate voices from field recordings for analysis. Compared to Demucs, Spleeter favors a lightweight TensorFlow-based toolchain and an MIT license, while Demucs often targets state-of-the-art perceptual quality via different model architectures.
Three capabilities that set Spleeter (Deezer) apart from its nearest competitors.
Which tier and workflow actually fits depends on how you work. Here's the specific recommendation by role.
Spleeter (Deezer) is useful when one person needs faster output without adding a complex workflow.
Spleeter (Deezer) should be tested for collaboration, quality control, permissions and repeatable results.
Spleeter (Deezer) is worth buying only if the pilot shows measurable time savings or quality gains.
Current tiers and what you get at each price point. Verified against the vendor's pricing page.
| Plan | Price | What you get | Best for |
|---|---|---|---|
| Free | Free | Full feature set, run locally; performance depends on your hardware. | Developers, hobbyists, researchers needing local separation |
| Enterprise (Self-hosted) | Custom | Scale depends on your infra; no vendor SLA from Deezer. | Organizations needing on-premise legal control and scalable processing |
Scenario: A small team uses Spleeter (Deezer) on one repeated workflow for a month.
Spleeter (Deezer): Free ·
Manual equivalent: Manual review and execution time varies by team ·
You save: Potential savings depend on adoption and review time
Caveat: ROI depends on adoption, usage limits, plan cost, output quality and whether the workflow repeats often.
The numbers that matter — context limits, quotas, and what the tool actually supports.
What you actually get — a representative prompt and response.
Choose Spleeter (Deezer) over Demucs if you prefer a lightweight, MIT-licensed TensorFlow tool with simple CLI and Docker integration.
Head-to-head comparisons between Spleeter (Deezer) and top alternatives:
Real pain points users report — and how to work around each.