Workshop for advanced users: The Fluid Corpus Manipulation project (FluCoMa)
Data mining through machine listening and machine learning.
March 14 – 18, 2022, 10:00 – 17:00
Location: Notam in Oslo
Price: 3000 or 5000 NOK for all academic staff in 50% positions or more, free for all others. Course participants who take the course for free must pay a deposit of NOK 300 which is refunded at the start of the course. The course is produced in collaboration with BEK – Bergen Centre for Electronic Arts.
The Fluid Corpus Manipulation project (FluCoMa), is a highly advanced tool for sound analysis, machine learning and sound synthesis in Max, Pd and SuperCollider. FluCoMa is a comprehensive sound processing tool, and can be used in everything from sound design and composition, to interactive installations and research. The workshop will include a mix of practical work, training and lectures.
FluCoMa is available for Max, Pd and SuperCollider. Max and SuperCollider will be the main tools in this workshop. Participants who use Pd will receive individual tutoring for this. Participants should preferably have a good, basic understanding of Max or SuperCollider to participate, but people with good skills in PD will also benefit from the workshop.
FluCoMa is a library for signal processing, machine learning and concatenative sound synthesis.
FluCoMa establishes new ways of exploring large collections of sound and gestures within music and sound design, by bringing breakthroughs of signal decomposition DSP and machine learning to the toolset of composers, artists and creative coders.
This allows groundbreaking sonic research into an unexploited area: the manipulation of large sound corpora. Indeed, with access to, genesis of, and storage of large sound banks now commonplace, novel ways of abstracting and manipulating them are needed to mine their inherent potential.
Three degrees of manipulations are set to be explored: (1) expressive browsing and descriptor-based classification, (2) remixing, component replacement, and hybridisation by concatenation, and (3) pattern recognition at component level, with interpolating and variation-making potential.
These novel manipulations will yield new sounds, new musical ideas, and new approaches to large corpora.
FluCoMa can be used in areas such as sound design, composition, interactive installations and research. The workshop will mainly be devoted to presenting the technology and giving participants the chance to use it through practical sessions, so that they can acquire the skills needed to create compositions, installations, digital instruments and their own creative tools.
More information about FluCoMa: https://www.flucoma.org/
Plan for the workshop
– introduction to FluCoMa
– basic MLP classifier
– basic MLP regressor
– basic slicing, description, 2D plotting and kdtree query
– adding a 3rd dimension and UMAP
– advanced/bespoke time series
– personal descriptors
– high d query (audio or control)
– mlp parameter
Read a detailed plan here.
Pierre Alexandre Tremblay
Pierre Alexandre Tremblay is one of the leading international experts in developing digital signal processing and machine learning for composition and sound design, and is one of the main developers of FluCoMa.
He teaches composition and improvisation at the University of Huddersfield in England, where he currently works on the FluCoMa project. He has also collaborated on a variety of projects, playing bass guitar and manipulating sound. He is a member of ars circa musicæ, de type inconnu, light.box, and Splice, blending from noise to free jazz to contemporary music.
He joined the University of Huddersfield in 2005, where he is now Professor of Composition and Improvisation. He previously worked in popular music as producer and bassist, and has a keen interest for creative coding.
James Bradbury is a creative-coder from Perth, Western Australia currently living in Northern England. In his practice he often uses the computer to organise audio-samples through machine listening and learning. He finished his PhD at the University of Huddersfield in 2021 under the supervision of Alex Harker and Steven Jan. James Bradbury currently work as a Research Fellow in Creative Coding on the FluCoMa project.
Ted Moore is a composer, improviser, intermedia artist, and educator. He holds a PhD in Music Composition from the University of Chicago and is currently serving as a Research Fellow in Creative Coding at the University of Huddersfield, investigating the creative affordances of machine learning and data science algorithms as part of the FluCoMa project.? His work focuses on fusing the sonic, visual, physical, and acoustic aspects of performance and sound, often through the integration of technology.
Three part workshop
The workshop is divided into three parts:
Monday and Tuesday: introduction to FluCoMa.
Wednesday, Thursday: FluCoMa for advanced users.
Friday: This day is for presentations, advanced questions and answers, as well as extracting universal questions regarding data mining of sound corpora.
Participants who have never used FluCoMa before must start on Monday. Participants who have participated in previous courses in FluCoMa or who are experienced users can choose whether they want to repeat the basics on Monday and Tuesday, or if they want to start on Wednesday.
Time: 14 – 18 March 2022, 10:00 – 17:00
Price for all academic staff in 50% positions or more: NOK 5,000 for 5 days (Monday-Friday), NOK 3,000 for 3 days (Wednesday-Friday). Free for everyone else.
Teachers: Pierre Alexandre Tremblay, James Bradbury, Ted Moore
Number of seats: 12
Location: Notam in Oslo
Registration deadline: March 7
Registration is confirmed by payment via one of these Paypal links:
FluCoMa at Notam 2022, academic staff in 50% positions or more, 5 days (5000 NOK)
FluCoMa at Notam 2022, academic staff in 50% positions or more, 3 days (3000 NOK)
FluCoMa at Notam 2022, deposit (300 NOK)
After you have paid, you will receive an email from Paypal. This is your confirmation that you have a place on the course.
We will send you detailed information about start-up and further follow-up as we come closer to the course start.
For questions regarding the workshop, send an e-mail to firstname.lastname@example.org