muzoti

Your Personal Algorithmic Composer.

Each unique piece composed and played for your ears only!

muzoti

Midi out:
Auto play:
Random select instruments:
Spatialise:

Best experienced in Chrome browser...

read more

What, another AI music maker?

Muzoti is the fruit of a 27 year personal obsession into what makes music sound, well... musical.

Given that a music score can to some degree be reduced to a sequence of numbers, why is it that one sequence sounds good, yet another sequence of randomly generated notes sounds terrible?

As the years went by I constantly tested my theories. I wrote programs that generated 'music' according the organising principles that I observed. The more musical the result, the more sure I became that I was on the right track...

Unlike other generative music attempts, we're not using machine learning. Instead we are developing a solid basis for generative musical intelligence based on what we believe are universal organising principles of music and life. Our algorithms are style agnostic. Without such a solid framework, machine learning can only extract style information based on pre-comosed works. We believe machine learning should come second to a holistic theory of what music is and why it exists.

Join the Muzoti early access group - get upcoming news and improvements before anyone else!

Your email address

Hear Muzoti Compose

If the live composer above didn't work for you (best heard in Chrome), then listen to some pre-rendered tracks composed by Muzoti below.

All the these music were composed by the same algorithm, with macro parameters tweaked between them, such as overall tempo. The instruments to be used for each part were chosen manually. The notes, rhythms and chords were completely algorithmically determined.

The Heroes 2016 set was created especially for a giant puppet performance on the South Bank, London in August 2016 as part of the Rio Olympic Opening Celebrations. Sounding very different - this shows the variety that can be achieved by simply changing the instruments and a few macro parameters. The pieces were generated by running the program about 40 times, each time with different hand-picked instruments and finally selecting the best 23 pieces.

Who is responsible for Muzoti?

Founder

Dr. Jeremy Leach - Algorithmic composition PhD and the driving force behind Muzoti.

Advisory team

Julien Bouteloup - Startup veteran; technology and blockchain evangelist

Carles Andres - Full stack javascript expert; energy tech enthusiast

What's next for Muzoti?

As you've seen, Muzoti is currently operating as a compose-on-demand service and able to produce a real-time online stream of algorithmic music.

For situations where branded recognisable music is not necessary, Muzoti could be used in this way to provide continuous unique music for environments such as:

  • Spas and wellness centres
  • Hotel lobbies
  • Hold music

Coming soon - a real-time responsive algorithmic composition system

Being able to influence the music as it is being composed is the next stage. Not just macro parameters such as tempo and loudness, but actually having parametric control over the melodic forms and interplay bewtween instruments. This opens up a number of extra opportunities:

  • In the gaming industry - for unique realtime context sensitive music that changes as the player's mood and environment changes
  • Interactive Art installations - using sensors to influence the music in real-time
  • Aiding music composition and idea generation
  • DJing
  • Music education

Interested in what we're doing?

Get in touch. Email Jeremy at office@muzoti.com

With thanks to...

Our WebAudio playback makes use of, or has been inspired by the following open-source work:

  • FluidR3_GM Soundfont, the excellent WebAudio General MIDI instrument pack. Without this there would be no instrument sounds!
  • The article about the ground-breaking Touch Pianist gave us some ideas about how to emulate quieter piano sounds, notably by using low pass filters. Thanks!
  • Base64Binary, used to convert the soundfont into audio data.
  • reverbjs.org an excellent resource of impule responses and a javascript library to use them.