spleeter — Source Separation Library with Pretrained Models for Audio Processing

Overview

Spleeter is a Python-based library developed by Deezer for source separation in audio files, utilizing TensorFlow. It offers pretrained models for separating vocals, drums, bass, and other elements from music tracks, and is known for its high performance and speed, especially on GPU.

Key Features

Pretrained models for 2, 4, and 5 stems separation
High performance on the musdb dataset
Fast separation capabilities, up to 100x faster than real-time on a GPU
Command line and Python library usage
Installation support via pip and Docker

Use Cases

Audio professionals for separating vocals from伴奏 in music production
Researchers and developers for training custom source separation models
Educational purposes for understanding audio signal processing

Advantages

State-of-the-art pretrained models for various separation tasks
High separation performance and speed, especially with GPU acceleration
Flexibility in usage through command line or as a Python library

Limitations / Considerations

May require significant computational resources for training custom models
Pretrained models are specific to certain separation tasks and may not cover all use cases
License information is currently unknown, which could affect usage rights

librosa: A Python package for music and audio analysis, offering a range of functionalities but without pretrained separation models.
TensorFlow: An open-source machine learning framework that can be used for building custom source separation models from scratch.
Deezer Research: Other projects and research from Deezer that may be related to audio processing and machine learning.

Basic Information

GitHub: https://github.com/deezer/spleeter
Stars: 27,268
License: Unknown
Last Commit: 2025-08-20

📊 Project Information

Project Name: spleeter
GitHub URL: https://github.com/deezer/spleeter
Programming Language: Python
⭐ Stars: 27,268
🍴 Forks: 2,987
📅 Created: 2019-09-26
🔄 Last Updated: 2025-08-20

🏷️ Project Topics

Topics: [, ", a, u, d, i, o, -, p, r, o, c, e, s, s, i, n, g, ", ,, , ", b, a, s, s, ", ,, , ", d, e, e, p, -, l, e, a, r, n, i, n, g, ", ,, , ", d, e, e, z, e, r, ", ,, , ", d, r, u, m, s, ", ,, , ", m, o, d, e, l, ", ,, , ", p, r, e, t, r, a, i, n, e, d, -, m, o, d, e, l, s, ", ,, , ", p, y, t, h, o, n, ", ,, , ", t, e, n, s, o, r, f, l, o, w, ", ,, , ", v, o, c, a, l, s, ", ]