Twinthesis : Twitter Powered Synthesis

Twinthesis is a MaxMSP patch I designed to explore the ‘sound’ of twitter, in an attempt to sonify the human randomness being generated on the service. This post is a quick overview of the synthesis engine, as well as a quick video outlining the features and concepts behind the patch. You can then download the synthesiser as a Mac application.


The aims of this project, are to create a synthesiser capable of both additive and granular synthesis using live tweets to generate and manipulate the sound. The synthesiser currently calls twitter once every 30 seconds, so a new tweet is used to generate the sound every 30 seconds. The synthesis engine, has an element of performance to it, and can be used to create experimental music. An example of experimental music created by the synthesis engine is here:

Twinthesis by TheHarmonizer

A full scientific paper and report can be downloaded about twinthesis, detailing aspects of how the patch works, and certain constraints of the project in it’s current state. Please note, this synthesiser is still in development and can be considered an experimental BETA version as released below. On a Mac the sound defaults to the ‘core-audio built in output’ at the moment.

Download – Scientific Paper / Report

Download – Twinthesis Application (Mac OS X Only)

As always, I appreciate all your interest in this project and am more than happy to answer any questions you may have, either in the comments below or via the contact page. I am also willing to share the MaxMSP patches upon request.

Many Thanks once again for reading!

Author: Sam

iOS Developer & blogger

27 thoughts on “Twinthesis : Twitter Powered Synthesis”

    1. Hey there,

      Very cool patch. Interested in how you used Max to get Twitter data into your patch. I’m trying to do something similar but with Jitter, and the social API integration has been the most challenging part.

      Any thoughts or guidance on this subject would be greatly appreciated.

      Keep up the fantastic work.

  1. This is awesome! I will be using it and studying soon, I'm very interested on gathering the info from tweets of specified users

  2. Wonderful – I really like the idea. Unfortunately I live in the Windows world with Max. Are there plans to expand the idea further visual (?) perhaps the game engine of Blender? Great work – thanks!!!

    1. Thank you! Yes there are plans to expand the idea further (had to devote time to my degree more recently) but I have plans for several features, graphics being the main development focus! I will also be releasing a windows version soon!

  3. Hi, first off, great work man! I'm attempting to do a similar project dealing with video mixed with live twitter feeds. I was wondering if it would be possible to use the portions of your patch that pulls from the twitter feed? I have some experience with Max, but my solution for pulling and parsing data from a website is incredibly ugly and slow. Please shoot me an e-mail, thank you.

    1. I would also be very interested in the Max patch portion that is pulling the data from twitter (and possibly other open forums too?). I am running a series of lectures on media art and I find the approach very interesting but would like to be able to find alternative ways of interpreting the data.

  4. Amazing things here. I am very satisfied to look your post. Thanks a lot and I am having a look ahead to contact you. Will you kindly drop me a mail?

Leave a Reply

Your email address will not be published. Required fields are marked *