Improtech 2019 | 3nd day

Welcome coffee

Improvisation, Digital Intelligence and Cultural Heritage

Keynote talk: “From Digital to Human Intelligence in Music Understanding Research”

Xavier Serra (Universitat Pompeu Fabra, Spain)

We are able to develop AI algorithms that solve complex musical tasks, yet, we are unable to apply these powerful technologies to help understand and improve our own musical comprehension abilities. Our machines are rapidly becoming capable of “understanding” music, while we still use traditional and time-consuming educational methods for training people in the development of their basic musical skills, or for that matter, in the development of most cognitive-based human capabilities. As listeners, in order to make sense of a particular music listening experience, we identify relevant auditory cues and then piece the cues together into patterns that can be retained long enough for brain mechanisms to examine and create the impression of auditory objects. Music lovers that appreciate and comprehend a particular musical style are able to verbalize their cognitive experience after listening to a music piece of that style. In this talk, Serra proposes that by building on prior research from the fields of Music Cognition, Music Information Retrieval, and Music Education we should be able to develop tools and perceptual training methodologies with which to help a naive listener to understand and appreciate a music tradition to which they had no prior exposure. Given that computers will never be able to comprehend or feel for us, we should do our best to build systems that can help us with that.

“'Jazz Mapping': Thematic Development and Story Telling in Jazz Improvisation”

Dimitri Vassilakis (National and Kapodistrian University of Athens, Greece)

“Jazz mapping” is a multi-layer analytical approach to jazz improvisation based on hierarchical segmentation and categorization of segments, or constituents, according to their function in the overall improvisation. In this way, higher-level semantics of transcribed and recorded jazz solos can be exposed. In this approach, the knowledge of the expert jazz performer is taken into account in all analytical decisions. In this talk, the speaker applies this method to two well-known solos, by Sonny Rollins and Charlie Parker, and discusses how improvisations resemble storytelling, employing a broad range of structural, expressive, technical and emotional tools usually associated with the production and experience of language and of linguistic meaning. The choices of the experienced jazz improviser who has developed a strong command over the language and unfolds a story in real time are made explicit. Very similar to prose on a given framework, he/she utilizes various mechanisms to communicate expressive intent, elicit emotional responses, and make his/her musical “story” memorable and enjoyable to fellow musicians and listeners. Vassilakis also comments on potential application areas of this work related to music and artificial intelligence.

“Metrical Polyrhythms and Polytemporality in live Improvisation Settings”

Sami Amiris (American College, Athens Big Band, Greece), Antonis Ladopoulos (National and Kapodistrian University of Athens, American College, Greece)

Artistic explorations of two improvising musicians who dare to dwell on the fringe of polytemporality and metrical polyrhythmicity. The musicians forming the critically acclaimed Phos Duo, expand their horizons by each expanding on distinct metrical surfaces – thus each one has a different sense of the whole than the other, like observers with different speeds in the theory of relativity – while still listening to each other and interacting. To make this possible in practice, a new type of metronomic sequencer – like an individual mechanical conductor – is used, capable of handling different timings simultaneously and independently from channel to channel, but still controllable overall. The result is an exciting and challenging musical environment that the duo finds fruitful for the creation of new musical textures, both through-composed and improvised.

Coffee Break

“Polyphonic Conversations”

Peter Nelson (University of Edinburgh, UK)

“... what’s intriguing about the ... improvisation I do is that its more polyphonic than your average conversation ...” John Oswald, interview 2009.

In recent years, the technologies used in making music and other time-based arts have changed radically, engaging with communication networks that have settled into every crevice of our social realities. This circumstance changes even the notion of what we consider to be the social, as well as the ways in which we engage with artefacts, processes and institutions. This talk will explore the implications of current technologies for improvisation strategies, and will interrogate the conversation as an expanding and transforming discourse whose agents may come to include more than simply human selves.

“Disposable Music”

Georg Hajdu (Hochschule für Musik und Theater Hamburg, DE)

This presentation introduces the concept of real-time composition and composition as a dispositif in the sense of Foucault and Deleuze, defining it as a heterogeneous ensemble of pieces which together form an apparatus. The introduction situates the dispositif in the context of cultural developments, most notably its slow, but steady shift away from textualization in digital media. As musicians are adapting to ensuing cultural and, above all, economic changes, new music forms emerge, such as comprovisation or laptop performance, which rely to a lesser degree on fully-notated scores. Antithetically, the computer also allows the creation of “author-less” notated scores in real-time to be sight-read by capable musicians – a practice for which special software has been developed in recent years. Since these scores are not meant to be kept and distributed, they are ephemeral and, therefore, disposable. Examples are given to illustrate the interwovenness of this approach, where carefully selected narratives and dramaturgies make up for the inherent unpredictability of the outcome.

Game, Mobiles, Transducers

“The Dynamic Percussion System: A procedural music engine for video games”

Daniel Brown (Intelligent Music System, USA)

The Dynamic Percussion System is a software used in commercial video games like “Rise of the Tomb Raider” (2015) that composes procedural percussion music that adapts to game action in real time. While the use of procedural music in video games is an exciting development, there are many unanswered questions and issues surrounding it. How does it fit into the traditional workflow of professional composers and sound designers? What choices do such people have when “authoring” procedural music? How does it interact with precomposed game music? The Dynamic Percussion System comprises both an in-game playback system and an authoring tool to be used by composers and sound designers. The design of the authoring tool – its user interface and functionality –addresses these questions. It offers one model of how the new techniques of generating procedural music can be adapted into the traditional methods used in commercial development. It has also motivated new techniques for generating, implementing, and interpreting game music.

“Improvisation with Motion Sensors and Live coding: Combining Dance and Instrumental Improvisation”

Ioannis Zannos (Ionian University, Greece)

This workshop introduces techniques of improvisation with wearable movement sensors combined with live coding. Movement sensors based on IMU (Inertial Measurement Units) are used to measure the movement of a performer. The motion data is transmitted to a computer over WiFi, and live coding is used to control the generation of sound in SuperCollider and graphics on the Godot Gaming engine. The workshop shows how to use the sc-hacks library in order to program and modify the response of the system to sensor data. Techniques for sending the control data to remote locations over the internet are shown. This enables joint performance from several different remote locations. At each location the data sent from all performers is used to synthesize the audiovisual performance.

“Composing 'musiques mixtes': acoustic spaces, improvisation and gestures”

Lara Morciano (composer, Italy), Jose-Miguel Fernandez (composer, Ircam, France)

The workshop introduces some possibilities of real-time interaction between instruments and electronics, explored through a device which uses the hands movements of the performer to control and synchronize various sound processes. This presentation will mainly focus on the “Philiris” composition for piano, motion capture and transducers. The different sections of the piece will be presented from a technical and compositional point of view. Examples of the interaction between the real and the virtual piano – linked to motion capture, sound synthesis and real time treatments through transducers within the piano – and the notion of “double” will be presented. It will also show examples of Antescofo, a score follower and programming language software, for the creation of real-time processes and interaction in relation with the notions of acoustic space, improvisation and gestures.

Jaap Blonk, Hervé Sellin, Georges Bloch

“Paris bout à bout”
Cine-concert on the film “Bout à Bout” by Nurith Aviv. Voice, piano, OMax and DYCI2 computer systems

“Bout à Bout” [End to End] is a series of sequence-shots filmed by Nurith Aviv in 1993, showing a poetic wander around Paris (in the subway, in particular). The film will be punctuated with still photographs shot by the artist from her window. This ensemble allows using live music and sound in a somehow similar way as for regular screenings with live music [ciné-concerts], but also as a source for real-time film-editing achieved by the improvising machines. Direct sound from the movie, Jaap Blonk’s sound poetry, Georges Bloch’s voice and Hervé Sellin’s piano mix with OMax and DYCI2 improvisation systems are brought together in an ambiguous hybridization.
Jaap Blonk (voice), Hervé Sellin (piano), Georges Bloch (voice, DYCI2 & Omax improvisation systems), Images by Nurith Aviv

Mark Bokowiec and Julie Wilson-Bokowiec

“HEXIS”
For Kinaesonics, Bodycoder System and soloist

“Hexis” is a Greek word that is important in the philosophy of Aristotle. It stems from a verb related to “possession” and is typically translated in modern texts as referring to a “state” of being. Sachs translates and refines the meaning as “active condition” and argues that hexis refers to an active disposition that becomes a deeply embodied “state” as opposed to a passive or surface action/interaction. For Bokowiec and Wilson-Bokowiec hexis describes a way of being in performance. It is a deeply embodied active condition that operates at the core of an interactive phenomenology and fluid materiality that is central to their particular brand of improvisational kinaesonics. Their piece “Hexis” (2019) is thematically inspired by and draws upon Heraclitus’ older pre-Socratic river-analogy: “one cannot step twice into the same river” and for “as they step into the same rivers, different and (still) different waters flow upon them”. These ancient philosophical fragments (attributed to Heraclitus via Didymus and Plutarch) configure man’s state of being in relation to and as a part of a substance of nature. The river is an example of that which preserves structural identity while undergoing constant change of content, and it is this conflagration of unity/divergence, formation and dissolution, of the catching, touching and trickling away of substance, that has informed the compositional architecture and underlying aesthetic principles of the piece. “Hexis” is the second work that seeks to explore the synergies between ancient motifs and Bokowiec and Wilson-Bokowiec’s own distinct form of sonic mediation, with the first being “PythiaDelphine:21” (2016) created with the kind support of Prof. Anastasia Georgaki and the National and Kapodistrian University of Athens and premiered at Animart in Delphi.
Mark Bokowiec (composer/MSP/electronics), Julie Wilson-Bokowiec (performer)

Marc Chemillier and Camel Zekri

“Gnawa Machine”
Augmented guitar, keyboard, DJazz and Le Cercle computer systems

This duet explores a dialogue between two machines. On one side, Marc Chemillier uses DJazz, a sofware belonging to the OMax family developed by IRCAM and EHESS (CAMS), which records on the fly the music played by a musician and improvises with it. The specificity of DJazz is to handle rhythm with various interfaces that allow agogical manipulation of the sequences generated by the computer. On the other side, Camel Zekri has developed his own electronic device in collaboration with the GRAME research center. It makes use of samples and is able to record them directly from an input signal. Thus, the two software environments will respond one to the other creating an amazing “mise en abyme” of machines. Some of the musical data injected into the process are taken from the Gnawa tradition which Camel Zekri belongs to.
Camel Zekri (augmented guitar, Le Cercle computer system), Marc Chemillier (keyboard, dJazz computer system)

Orestis Karamanlis and Giorgos Gargalas

“BitVox”
For beatboxing and laptop live electronics

“BitVox” is a work for beatboxer and live electronics running in the SuperCollider language. It makes use of audio event analysis and algorithmic beat manipulation in an effort to create a unified soundworld between the human beatboxer and his mechanical counterpart. It utilises a custom built SuperCollider class (the CuePlayer) allowing the organization of processes and musical material in bundles and their execution in sequential cues.
Orestis Karamanlis (laptop), Giorgos Gargalas (beatboxing)

Steve Lehman and Jerôme Nika

“Silver Lake Studies”
Saxophones, live electronics, DYCI2 computer system

“Silver Lake Studies” is the first presentation of a research project that Steve Lehman and Jérôme Nika have been conducting together for more than a year. Centered around the DYCI2 program, this collaboration focuses on the integration of scenarios in music generation processes, and on the dialectic between reactivity and planning in interactive human-computer music improvisation. During their experiments, they explored the creation of generative processes that can adapt to constantly evolving metrics, the development of real-time spectral enrichments of saxophone improvisation, and “spectral chord changes” as the basis for melodic improvisations. In this work, the orchestral masses from the contemporary classical repertoire meet voices from experimental rap and hip-hop.
Steve Lehman (saxophones, electronics), Jérôme Nika (DYCI2 computer systems)

Click here to read more about the participants