A library for audio and music analysis, detecting events like onsets, pitch, tempo, and beats in audio signals.
aubio is a library for audio and music analysis that detects events in audio signals, such as note onsets, pitch frequencies, and rhythmic beats. It provides a suite of algorithms for segmenting sound files, performing pitch detection, tracking tempo, and extracting features like MFCCs. The library solves the problem of programmatically analyzing and labeling audio content for applications in music technology and sound processing.
Audio engineers, music technologists, and developers working on music information retrieval, audio feature extraction, or real-time sound analysis applications.
Developers choose aubio for its comprehensive set of proven audio analysis algorithms, lightweight performance, and cross-platform compatibility. Its straightforward API and command-line tools make it accessible for both research and production use in audio processing pipelines.
a library for audio and music analysis
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Offers multiple methods for onset detection, pitch estimation, and tempo tracking, as listed in the key features, providing flexibility for different audio analysis tasks.
Compiles on Linux, Mac, Windows, and iOS, per the build instructions, ensuring it can be used in diverse development environments.
Includes tools like aubioonset and aubiopitch for direct audio analysis without coding, making it accessible for quick experiments or scripting.
Features a detailed manual and developer docs hosted on aubio.org, with badges indicating active documentation maintenance.
Requires compiling from source or managing C dependencies, as noted in the build instructions, which can be a barrier for users unfamiliar with build systems.
The library's name acknowledges inherent errors in audio analysis, and its traditional algorithms may not match the precision of modern machine learning-based approaches.
Focuses solely on audio analysis and labeling, lacking built-in features for synthesis or advanced editing, which might require integrating additional libraries.