Spleeter
Source: https://github.com/deezer/spleeter
Structured fields
{
"archived": false,
"checked_at": "2026-10-17T00:00:00.000Z",
"usefulness": "Deezer's fast source-separation library with pretrained 2-, 4- and 5-stem models (vocals/accompaniment, +drums/bass/other, +piano), evaluated on MUSDB; older baseline now surpassed by newer separators such as Demucs.",
"source_type": "repository",
"availability": "available",
"source_owner": "Deezer",
"next_check_at": "2026-10-24T00:00:00.000Z",
"upstream_updated_at": "2026-06-18T15:22:32.000Z"
}Evidence claims
- [upstream_claim] README: separation into 2 stems (vocals/accompaniment), 4 stems (vocals/drums/bass/other) and 5 stems (plus piano); CLI example: spleeter separate -p spleeter:2stems -o output audio_example.mp3; 2.1.0 introduced breaking CLI changes and dropped the dedicated GPU package. · locator: README, About and usage source
- [upstream_claim] README: 2 and 4 stems models reported to have high performance on the MUSDB dataset and used by third-party products such as VirtualDJ stems. · locator: README, performance notes source
- [sourced_fact] README License: Spleeter code is MIT-licensed; GitHub: not archived, last push 2026-06-18. · locator: README, License source
- [upstream_claim] Research review verdict at export: reviewed (as of 2026-10-17). Recorded by the research pipeline, not a hub review outcome; evidence.review stays unreviewed until an independent reviewer records one.
Rights
| component | license | hosting | redistribution | modification | training | Evidence and notes |
|---|---|---|---|---|---|---|
| code | MIT | permitted | permitted | permitted | unknown | MIT code per README. |
| model_weights | unknown | unknown | unknown | unknown | unknown | Pretrained model terms not separately stated in the excerpt read; training data not reviewed. |