AUBIOONSET(1)

🌸 AUBIOONSET 🌸

ok so this is a page from the aubio User's manual (section 1 of the man pages, bestie, very official) and it's all about aubioonset, the little command-line icon that finds the moment a sound starts. literally the beginning of everything. main character energy fr fr 💅

---

✨ NAME ✨

aubioonset — a command line tool to extract musical onset times 🎀

---

🦋 SYNOPSIS 🦋

aubioonset source
aubioonset [[-i] source] [-o sink]
           [-r rate] [-B win] [-H hop]
           [-O method] [-t thres]
           [-T time-format]
           [-s sil] [-m] [-f]
           [-j] [-N miditap-note] [-V miditap-velo]
           [-v] [-h]

---

💖 DESCRIPTION 💖

okay so picture this: aubioonset is out here trying to detect onset times — and an onset is literally just the beginning of a discrete sound event in an audio signal. that's it, that's the job. finding the exact moment the vibe changes. 🫶

when you start it with an input source (-i/--input), the detected onset times get printed right there on the console, in seconds. no notes, no drama, just numbers bestie. ✨

but when you start it without an input source — or when you slide the jack option (-j/--jack) into the chat — aubioonset boots up into jack mode. different era, different energy. 🌈

---

💅 OPTIONS 💅

so this program keeps it classic and follows the usual GNU command line syntax, which means long options start with two dashes (--). here's the full rundown, no skips, no cuts:

• -i, --input source — run the analysis on this audio file. most uncompressed and compressed formats are supported, depending on how aubio was built. yes, that's a real caveat. build-dependent. 🙈
• -o, --output sink — save the results into this file. the file gets created on the model of the input file, and onset times are marked with a short wood-block-like sound. cute little knock for every hit. 🎵
• -r, --samplerate rate — fetch the input source, resampled at this sampling rate. the rate goes in Hertz as an integer. if it's 0, the sampling rate of the original source is used. defaults to 0.
• -B, --bufsize win — the size of the buffer being analyzed, aka the length of the window used for the spectral and temporal computations. defaults to 512.
• -H, --hopsize hop — the number of samples between two consecutive analyses. defaults to 256.
• -O, --onset method — the onset detection method to use. peep the ONSET METHODS section below for the lineup. defaults to 'default'.
• -t, --onset-threshold thres — sets the threshold value for the onset peak picking. values typically live in the range [0.001, 0.900]. lower threshold values imply more onsets get detected; raising this threshold should reduce the number of incorrect detections. defaults to 0.3. tune responsibly. 🔥
• -M, --minioi value — sets the minimum inter-onset interval, in seconds — so, the shortest interval allowed between two consecutive onsets. defaults to 0.020.
• -s, --silence sil — sets the silence threshold, in dB, under which the onset will not be detected. a value of -20.0 would eliminate most onsets but the loudest ones; a value of -90.0 would select all onsets. defaults to -90.0. 📉
• -T, --timeformat format — sets the time format (samples, ms, seconds). defaults to seconds.
• -m, --mix-input — mix the source signal into the output signal before writing to the sink. blend era. 🫶
• -f, --force-overwrite — overwrite the output file if it already exists. no hesitation.
• -j, --jack — use Jack input/output. heads up: you will need a Jack connection controller to feed aubio some signal and to listen to its output.
• -N, --miditap-note — override the note value for the MIDI tap. defaults to 69.
• -V, --miditap-velop — override the velocity value for the MIDI tap. defaults to 65.
• -h, --help — print a short help message and exit. the classic escape hatch.
• -v, --verbose — be verbose. say more. 💬

---

👑 ONSET METHODS 👑

the available methods are, in this exact order, and i'm not skipping a single one:

default

default distance, currently hfc.

Default: 'default' (currently set to hfc)

energy

energy based distance.

this function calculates the local energy of the input spectral frame.

hfc

High-Frequency content.

this method computes the High Frequency Content (HFC) of the input spectral frame. the resulting function is efficient at detecting percussive onsets. 🥁

Paul Masri. Computer modeling of Sound for Transformation and Synthesis of Musical Signal. PhD dissertation, University of Bristol, UK, 1996.

complex

Complex domain onset detection function.

this function uses information both in frequency and in phase to determine changes in the spectral content that might correspond to musical onsets. it's best suited for complex signals such as polyphonic recordings.

Christopher Duxbury, Mike E. Davies, and Mark B. Sandler. Complex domain onset detection for musical signals. In Proceedings of the Digital Audio Effects Conference, DAFx-03, pages 90-93, London, UK, 2003.

phase

Phase based onset detection function.

this function uses information both in frequency and in phase to determine changes in the spectral content that might correspond to musical onsets. it's best suited for complex signals such as polyphonic recordings.

Juan-Pablo Bello, Mike P. Davies, and Mark B. Sandler. Phase-based note onset detection for music signals. In Proceedings of the IEEE International Conference on Acoustics Speech and Signal Processing, pages 441444, Hong-Kong, 2003.

specdiff

Spectral difference onset detection function.

Jonhatan Foote and Shingo Uchihashi. The beat spectrum: a new approach to rhythm analysis. In IEEE International Conference on Multimedia and Expo (ICME 2001), pages 881884, Tokyo, Japan, August 2001.

kl

Kulback-Liebler onset detection function.

Stephen Hainsworth and Malcom Macleod. Onset detection in music audio signals. In Proceedings of the International Computer Music Conference (ICMC), Singapore, 2003.

mkl

Modified Kulback-Liebler onset detection function.

Paul Brossier, ``Automatic annotation of musical audio for interactive systems'', Chapter 2, Temporal segmentation, PhD thesis, Centre for Digital music, Queen Mary University of London, London, UK, 2006.

specflux

Spectral flux.

Simon Dixon, Onset Detection Revisited, in ``Proceedings of the 9th International Conference on Digital Audio Effects'' (DAFx-06), Montreal, Canada, 2006.

---

🌟 SEE ALSO 🌟

go give these icons a follow, they're all in the same friend group:

aubiopitch(1), aubiotrack(1), aubionotes(1), aubioquiet(1), aubiomfcc(1), and aubiocut(1).

---

🎀 AUTHOR 🎀

this manual page was written by Paul Brossier <piem@aubio.org>. permission is granted to copy, distribute and/or modify this document under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. 💜

---

okay that's the whole page, you ate that fr fr 💅✨ this was aubio 0.4.9, dated 29 March 2025 — mwah, go detect some onsets bestie 🦋🍓