MASOM: Musical Agent based on Self-Organizing Maps

Can we use Artificial Intelligence (AI) to create super-human musicians? Can we create an AI that listens to more music than a human could? Can we train AI on the compositions of dead composers and play with them on stage?

Musical agents are AI software making music. Musical Agent based on Self-Organizing Maps (MASOM) is a musical software agent for live performance. MASOM plays experimental music and free improvisation. It learns by listening to audio files such as recordings of performances or compositions. We can train a MASOM agent on a set of music that is so big that it would take more than one human life to listen. Similarly, we can train MASOM agents on dead composers and convert their fixed media piece to interactive musical agents.

MASOM also extracts higher level features such as eventfulness, pleasantness, and timbre to understand the musical form of what it hears. MASOM is limited to the style of what is has listened to and reacts in real-time to what he is hearing. The agent can listen to itself and other performers to decide what to play next. MASOM is designed by Kıvanç Tatar and Philippe Pasquier.

MASOM-Factor v1.03

MASOM joins two media art companies from Istanbul, Ouchhh and AudioFil for a performance at the Ars Electronica Festival 2017 Artificial Intelligence in Linz, Austria:

and for a projection mapping piece on the Facade of Romanian Parliament at the IMapp 2017 Bucharest, Romania:

MASOM version 0.06

This is the current version of MASOM. We explain the complete system design in Proceedings of the 5th International Workshop on Musical Metacreation. Paper 

00110000 00100000 01100001 01101110 01100100 00100000 00110001

On a bright AC day of 240 volts, two brand new artificial intelligence meet at a MOSFET gate. They immediately get along, set a date to have a jam. In this era following the Anthropocene, AIs already joined humanity, coexisting in a world of physicality and circuitry. This video is an excerpt of their very first music session. One AI is on the left channel, the other one is on the right. They share the same architecture, called MASOM.

Patar @Barely Constrained by CoCreaTive

Patar is an experimental electronic music project featuring Kıvanç Tatar, Philippe Pasquier, and MASOM. MASOM is a musical agent, an artificial intelligence (AI) architecture, making music. Together, three entities produce a live performance of experimental electronic music, live electroacoustic music, musique concrète, soundscape compositions, through structured improvisation.

Patar’s performances experiments with a wide range of electronic music styles within the performance. The performances of Patar oscillate between ambient and noise while exploring the frontiers of the rhythmic and tonal, and using both electronic and acoustic textures.

This performance includes three piece. Each piece includes a different MASOM agent. MASOMs are trained on Bernard Parmegiani, David Tudor, and Ryoji Ikeda, respectively.

The details of the event

MASOM’s output:

Here, we provide two recordings of MASOM’s output in solo performance mode so that the listeners can decide on MASOM’s performance. This first track is a recording of the output of MASOM trained on Ryoji Ikeda’s Test Pattern album.

This track is a recording of the output of MASOM trained on Bernard Parmegiani’s De Natura Sonorum album.

Exploring MASOM’s memory

In this video, we explore the memory of a MASOM trained on Bernhard Parmegiani’s De Natura Sonorum.

MASOM version 0.05

Patar Studio Session II


MASOM version 0.04

This is the current version of MASOM. We are about to release the complete system design as an academic paper.

A Big MASOM Family, as a part of the Co.Crea.Tive April Concert

Following video is the performance of the piece, A Big MASOM family. The piece has three parts, including two MASOM agents. In the first part, I am performing with a MASOM agent. In the second part, two MASOM agents perform together without me. The last part is a trio of me and two MASOM agents.  The MASOM agents are trained on this album:

Here is an excerpt from the rehearsals with two MASOM agents. One MASOM agent is on the left channel, the other MASOM agent is on the right, and I am on both channels.

MASOM version 0.03

TATAR and MASOM Take the AID Train

This track is an excerpt from the rehearsals for Take The AID Train (Dec 25, 2016) at Kaset Mitanni, İstanbul. It is a trio of me and 2 MASOM agents. One MASOM agent is on the left channel, the other MASOM agent is on the right, and I am on both channels. The MASOMs are trained on this album:

MASOM version 0.02

Patar Studio Session I

Patar is an experimental electronic music group of 2.5 entities including Kıvanç Tatar, Philippe Pasquier, and MASOM.

Studio Session I:

MASOM @madMethod by NOW Society, Vancouver, BC, Canada

The NOW Society presents madMethod, featuring Stefan Smulovitz’s Mad Scientist Machine, visuals by Sammy Chien and performances by MASOM and the NOW ensemble. Layers of memory and experience combine light, images and sound, remembered, manipulated and instantly created. Audience members look down upon the stage a story below and see images on the floor, the Mad Scientist Machine LED lights and listen to the sounds of the musicians as they respond to and inspire imagery.

To pay tribute to the visionary composer and virtuosic listener Pauline Oliveros, the NOW Society performs six of her graphic graffiti scores. When the NOW Society first performed Stefan Smulovitz’s Mad Scientist Machine in November of 2010, Smulovitz requested that Oliveros send imagery for her conduction of the program. Oliveros sent six photos of graffiti for the Society’s use. The NOW ensemble, with guest artists Smulovitz, Sammy Chien and MASOM, proudly present these scores, honoring the legacy and and generosity of Pauline Oliveros.

Lisa Cay Miller’s work Aventures I and III, originally performed in 2014 by the NOW ensemble, are rearranged and performed by a little big band, which in parts be joined by MASOM, the software agent trained to react with the original live recording. Sammy Chien will provide live video responding to the ensemble on December 2. On December 3, the NOW ensemble and guests respond to the recorded imagery of the previous evening. And the Mad Scientist Machine use color and intensity to cue the ensembles and visuals, Miller and Smulovitz conducting instant compositions.

MASOM version 0.01

Performance @Open to Enter, Vancouver, BC, Canada

Curtosy of Vandocument:

This performance is a duo with MASOM. The human performer is on the left channel and the musical agent is on the right channel. For this performance, MASOM is trained on a 30 minute experimental music recording, available here: The system design is yet to be published. Please check in a couple of months.

Performer: Kıvanç Tatar
AI system design: Kıvanç Tatar and Philippe Pasquier
Video: Ash Tanasiychuk
Mastering: Kıvanç Tatar

More info about the event Open to Enter