---
id: librosa
version: "1.0.0"
license: ISC
license_treatment: permissive
maintenance: active
---

# librosa — Python module for audio and music processing

License: permissive · Maintenance: active · Downloads: 13.6M/mo

## What it is and what it does

librosa is a foundational Python library for audio and music signal processing, designed to support music information retrieval (MIR) research and applications. It provides algorithms for feature extraction, spectral analysis, and music analysis tasks. The library depends on scientific computing foundations (numpy, scipy, scikit-learn) and uses numba for performance-critical code, plus soundfile and soxr for audio I/O and resampling.

Developers use librosa to extract features from audio (mel-spectrograms, chroma, MFCC, onset detection), analyze musical structure, and build systems that understand or process music. It's production-stable (Development Status 5), actively maintained with a recent release, and widely adopted in music technology and research.

Use it for:

- Extract mel-spectrograms and other acoustic features from audio files for machine learning models.
- Detect musical onsets, beats, and tempo to analyze rhythm and timing in recordings.
- Compute chroma features and harmonic/percussive separation for music analysis and transcription tasks.
- Build music recommendation or similarity systems using librosa feature vectors.
- Analyze audio for speech processing, environmental sound classification, or audio event detection.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

librosa provides audio and music signal processing algorithms and tools for building music information retrieval systems in Python.

Yes. librosa is production-stable, actively maintained, has low install friction, carries a permissive license, and is the standard Python library for audio and music signal processing. Install it if you need to work with audio features, music analysis, or MIR tasks. No known security vulnerabilities.

## Install

pip install librosa
uv add librosa
poetry add librosa

## Installing librosa

Before you install: Low friction: pure Python wheel available, active maintenance with release 3 days ago, 11 runtime dependencies all widely used (numba, numpy, scipy, scikit-learn, joblib, soundfile, pooch, soxr, decorator, lazy_loader, msgpack).

License in practice: ISC license is permissive, allowing commercial and private use with minimal restrictions—suitable for most projects.

Quickstart:

pip install librosa

import librosa
y, sr = librosa.load('audio.wav')
S = librosa.feature.melspectrogram(y=y, sr=sr)

Requires Python 3.12 or later; soundfile and soxr may require system audio libraries depending on your platform.

Verify before relying:

- Whether soxr and soundfile have platform-specific system dependencies that could affect installation on Windows, macOS, or Linux.
- Performance characteristics when processing very large audio files or real-time streaming scenarios.

## Package facts

- License: ISC (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 13.6M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

audio signal processing python, music information retrieval, audio feature extraction, music analysis library, spectral analysis audio, audio waveform processing, music mir tools, audio-processing, music-analysis, signal-processing

[View on SkillFed](https://skillfed.io/packages/librosa) · [View on PyPI](https://pypi.org/project/librosa/)