Isotonik Studios Logo

Out of stock

Factorsynth 2

Machine Learning in Ableton Live


(6 customer reviews)

Factorsynth is a new kind of musical tool. It uses machine learning techniques to decompose any input sound into a set of temporal and spectral elements.

By rearranging and modifying these elements you can do powerful transformations to your clips, such as removing notes or motifs, creating new ones, randomizing melodies or timbres, changing rhythmic patterns, remixing loops in real time, creating complex sound textures…

Developer: JJ Burred
Compatibility: Ableton Live 10/11 Suite or Ableton Live 10/11 Standard with MaxforLive Installed
OS: Mac & Windows
Min. Requirements: MaxforLive 7 is Essential

Out of stock




Factorsynth is a Max For Live device created by J.J. Burred that uses a data analysis algorithm called matrix factorization to decompose any audio clip into a set of temporal and spectral elements. By rearranging and modifying these components you can do powerful transformations to your clips, such as removing notes or motifs, creating new ones, randomizing melodies or timbres, changing rhythmic patterns, remixing loops in real time, applying effects selectively only to certain elements of the sound, creating complex sound textures…

After 2 years of the initial release comes Factorsynth 2, the first major update. Following many user suggestions and requests, version 2 is an even more versatile yet easier to use device, with a simplified workflow and numerous new features. It is now possible to individually pan the components, allowing to do things such as upmixing a mono clip to stereo. Another powerful new feature is the quantized shifting of the components, which allows changing the rhythmic structure of riffs and drum loops. A second, alternative decomposition algorithm is available, as well as a more detailed control of the playback region.

A restricted Demo version of Factorsynth is available here – FACTORSYNTH 2 DEMO

How does Factorsynth work?
Factorsynth is based on a modified version of an algorithm called Non-Negative Matrix Factorization (NMF). Simply put, NMF can automatically extract interesting patterns from data. It has been used in fields such as computer vision and movie recommendations. I had to heavily adapt and tweak it in order to meet the real-time needs of music production.

Can Factorsynth remove a full voice/instrument from a mix?
That’s unlikely, unless your voice or instrument plays only a few sustained notes, with no effects and no vibrato. Factorsynth can extract interesting sound events, such as individual notes, attack noises, impulses or rhythmical structures (watch the demo video to get an idea), but it’s not aimed at separating full instruments. That’s the job of source separation, which is a harder thing to do! On the other hand, you can often nicely separate drum sets and individual drum instruments (kick drums, hi-hats, snares…). For more details about this, check out this blog post.

Can Factorsynth’s parameters be controlled by Live’s automation envelopes or MIDI mappings?
Most of them can (factorization buttons, matrix operation buttons, solo buttons, number of components, all levels and pan sliders, shifting buttons, transport controls and quantization). Also, the first 8×8 buttons on the switchboards are MIDI-assignable. The ones that cannot be controlled by Live are: analysis parameters, clear panel, export panel, “snap” and “all” buttons on the output waveform display, reset and threshold buttons and individual buttons on the switchboards corresponding to component numbers higher than 8.

Can Factorsynth be used with standalone Max?
Yes, starting from version 1.5 it is possible to use them with the standalone Max application, without Ableton Live. There is a bit of patching needed to set it up. For details, you can take a look at the “Usage from Max” section in the manuals.

6 reviews for Factorsynth 2

  1. Jamie (verified owner)

    A super Max tool that is different and spectacular. You can break a sound loop into multiple granular parts and rearrange it in almost infinite combinations. You could get lost for days. Top notch work. I highly recommend it for music and sound design. It’s a secret weapon for creativity!

  2. duneglow

    Amazing device. Looking very much forward to it’s future. Would love it if extracted elements could be made to have their own individual mixer channels, so they could be processed independently.

  3. Kieron Johnson (verified owner)

    Factorsynth is absolutely essential. There is NOTHING else like this out there, and for that it’s worth its weight in gold to anyone looking for unheard sounds. It’s still surprising me after a month of heavy use. I can’t think of the last time a piece of software had me this excited.

  4. Salvador España Ramirez (verified owner)

    Top M4Live device. Worth every euro spent. Thanks to J.J. Burred an Isotonik Studios for this jewel.

  5. moritz meschonat (verified owner)

    I bought Factorsynth and Iota today, explored them both playfully and wow… these are creativity tools like I was looking for without even suspecting it.
    Conclusion: unimagined possibilities for little money. buy.

    in my first (simple!) test i had “factorized” a 4 bar tape recording in Factorsynth. i faded out rhythmic elements, bass and piano. what was left were parts of the solo instrument with an interesting sound aesthetics. exported and with a portion of valhalla a really good start. i imported the new file into iota and created a rhythmic ambient chord accompaniment with three zones.
    both tracks together a great foundation. let’s see where i end up with it…

  6. David Rowe (verified owner)

    Factorsynth is really intuitive. An outside the box idea and I’m loven’ it! Thank you JJ for such a cool tool! 5 STARS!
    Factorsynth is the rumbling underpinning loop for this piece. I bounced a 2 bar loop, brought it into LIVE on two separate tracks and attached an Auto Filter and a Filter Delay and drove their parameters with an LFO. Each slightly different.
    I called it Shufflin Wub because of the sound created.

Add a review


v2.3 – 20/1/2021

– Version for compatibility with M1 Mac

v2.2 – 27/7/2021

– first demo version
– several bug fixes in shifting without quantization
– factorize buttons blink when analysis parameters changed
– single click to jump playback position
– fixed crash when hot-swapping presets

v2.1 – 12/2/2021

– Added triple meter quantizations
– Added recursive double/half tempo buttons
– Control of snap quantization by main quantization parameter
– Added beat number control
– Fixed host mode sync issues in Max 8.1.7 and later versions

v2.0 – 10/7/2020

– Avoid clicks when shifting components
– Avoid gain boost in host mode
– Bugfix: changing global levels bypassed the mute buttons
– Fixed some audio dropouts when moving dials
– Fixed memory allocation crashes on Mac OS Catalina
– Fixed crash when clicking on operation buttons after changing component number
– Fixed info window annotations

v2.0b4 – 4/6/2020

– Double/half tempo buttons
– Display left-right output as 2 superimposed waveforms
– New colors
– Fixed some undo issues
– Bugfix: display correct gain after shifting
– Normalize temporal shape displays if using algorithm B
– Windows bugfix: was not re-factorizing when reloading
– Bugfix: make sure output is zero if no components are selected
– Avoid loop clicks
– Bugfix: reset loop length after factorizing new master sound

v2.0b3 – 14/5/2020

– added zoom button
– increase the maximum gain to 16 dB
– Bugfix: show cancel button while exporting files
– Fixed crash when changing number of components while factorizing
– Fixed text overruns in Windows
– Fixed MIDI parameter names

v2.0b2 – 29/4/2020

– Fixed Max interface
– Some annotation fixes
– Bugfix: select right buffer for bouncing
– Updated manual to v2

v2.0b1 – 27/4/2020

– Import master sound by drag and drop
– Instant loading of warped or transposed tracks (instead of real-time recording)
– Playback and looping is now controlled from the device (solves sync problems)
– Loading from encrypted pack samples is now possible
– New decomposition algorithm available
– Iteration number is now a user parameter
– Panning controls for the components
– Possibility to select looping region on output waveform
– Temporal component shifting
– Playback cursor now covers all temporal shapes
– Thresholding of component amplitudes
– Factorizes right away when loading from presets
– Randomly add or remove individual components
– All-on operation buttons
– Hide cancel button while processing
– Avoid clicks when stopping
– Transport control and master file display on device window
– New output gain dial
– Button to open Max console
– Several small fixes is STFT computation
– Some memory optimizations in NMF computation

v1.5 – 2/5/2019

– Using the device from Max is now supported
– Multiple device instances are now supported
– Bugfix: keep saved switchboard connections after loading set with warped clips

v1.4 – 28/2/2019

– Manual MIDI assignment of the first 8×8 switchboard buttons.
– Automatic mapping of the switchboard for Launchpad and Push surface controllers.
– Allow to resize main interface window.
– Bugfix: correct instant launching of clips shorter than one beat.
– Prevent drag&drop to input and output waveforms.

v1.3 – 8/11/2018

– Added clear buttons.

– Avoid saving temporal and spectral elements to undo history (won’t be saved to presets).

– Bugfix in session view looping : was missing loop restart during heavy computations.

– Clear output waveform when changing tempo.

v1.1 – 27/8/2018

– Solved several looping problems in session view.

– Bugfix in element drawing when loading from preset or saved live set.

– Use fs_out.wav as default file name when bouncing to file.

– Bugfix: unwarped but transposed clips were not being recorded.

v1.0 – 10/7/2018

– Initial release



“Quite magical.” – ASK AUDIO

“A powerful instrument for breaking down audio loops in to timbral and spectral elements that can be remixed and exported in astoundingly cool ways.” – THE PRO AUDIO FILES


BONEDO – Review

JJSAUMA – Review


ASK AUDIO – Factorsynth Initial Release Feature 

MATRIXSYNTH – Factorsynth Initial Release Feature

SYNTHANATOMY – Factorsynth Initial Release Feature

SEQUENCER.DE – Factorsynth Initial Release Feature

IDM MAG – Factorsynth Initial Release Feature

REKKERD.ORG – Factorsynth Initial Release Feature

GEARNEWS.DE – Factorsynth Initial Release Future

MIDIFAN – Factorsynth Initial Release Feature

DJPROFILE.TV – Factorsynth Initial Release Feature



Our Products are generally provided as Zip file downloads which in all cases will need to be extracted and saved to your hard drive prior to installation. Details on the different file type provided and what to do with them are below.

.alp Files – These are Ableton Live Packs and will install directly into the Packs Section of the browser of Ableton Live. When you double click on them you will see a pop up window asking you to confirm that installation is ok as the files are described as “Legacy”. This is fine and you are ok to proceed, the packs are created using the format for an earlier version of Live and so remain compatible with the current version.

.amxd files – These are individual MaxforLive Devices and can be stored anywhere on your hard drive. We’d advise creating a folder for them and then adding that folder as a location within the Places section of Lives Browser.

Control Surface Scripts – Please refer to the individual User Guides provided with your download, an abridged version of the installation will be shown further down this page.

ClyphX Pro – Please refer directly to the manual and installation video for this specific product.

Novation Circuit Packs – These are provided as individual Sysex Files and as a complete bank file. All are to be installed using a version of the Circuit Editor and not via Novations Components back up utility.



J.J. Burred is an independent researcher, software developer and musician based in Paris.With a background in machine learning and signal processing, his work aims at developing innovative tools for music and sound creation, analysis and search.After earning a PhD from the Technical University of Berlin, he worked as a researcher at IRCAM and Audionamix, on topics such as source separation, automatic music analysis, sound classification, content-based search and sound synthesis. His current main activity concerns the exploration of machine learning techniques for new methods of sound analysis/synthesis aimed at musical creation.LINKS: WEBSITE – FACEBOOKYOUTUBESOUNDCLOUD 

View more products from this vendor