Live processing

The term live processing can include any type of signal processing that occurs in real-time. In this report we will use live processing to describe the concept where one musician is processing a live audio feed from another musician in real-time. The technical setup is not very remarkable in itself, but we also consider the type of interplay that occurs between the two musicians in this context to be an important and integral part of the concept live processing.

Early experiments

Øyvind and Carl Haakon had some experience in this field from their duo Little Soldier Joe (https://soundcloud.com/brandtsegg/sets/little-soldier-joe) where Øyvind is live processing the sounds from Carl Haakon’s acoustic drums. This setup was also brought in as a part of the T-EMP setup from the start. In addition Øyvind added input lines from Tone and Ingrid to his setup. This was a natural addition considering the aims for the research project and has been an integral part of the band´s sound and identity since the start.

After some time, the idea that it would be interesting to see what happened if more players could do live processing of the others grew, and during the workshop with Bent Sæther this was tested for the first time with Bernt Isak processing Ingrid and Trond processing Tone. Since neither Trond or Bernt Isak really had prepared their instruments specifically for this setup, the result was a bit chaotic and also a bit of a side track from what we were working on in that session. Even though, there were some elements that promised some quality output if worked on properly.

In line with the projects original aims we came to the idea that we should have a setup with multiple connections for live processing where several players were able to process input from several other players. This would potentially complicate the challenges of listening and communication where for instance identification could be even more deluded by multiple sources for one sound.

Working with jazz students

The first proper test of multiple players doing live processing at the same time was carried out as a 3 day workshop in a studio with 3rd year jazz students from The Department of Music. Most of the jazz students played their instruments acoustically (only a couple of guitarists had an amplifier and some effect pedals), but with microphones and listening through headphones. Trond, Bernt Isak and Øyvind (this time all three of us with instruments specifically prepared for live processing) got a multichannel live audio feed with the option of individual mixing to extract separate instruments or submixes.. Before the jazz students started to play, Øyvind and Carl Haakon played a short piece to demonstrate our use and perception of live processing.

Since this complex setup can quickly become chaotic if everyone is playing at the same time, we decided to play in duos for the first part of the workshop. Each student got to play once while Trond, Bernt Isak and Øyvind took turns as sound processor. Since this setup makes the interplay quite transparent, it proved very useful for teaching and analysis purposes as it was possible to decode the development of the interplay for those who were listening. It also became clear that live processing in this way can take on several forms. In terms of interplay it can vary from the two participants being individual players responding playing together in a duo constellation to a fusion where they blend into one player with a single expanded instrument.

There is also an interesting “power balance” that became very apparent in some of these duo constellations. Since the sound processor has no input of his own, he or she won’t be able to really do anything if the sound producer doesn’t produce sound. This limits the ability to take initiative with first impulses. On the other hand, the sound producer has no control over what kind of processing is applied to his or her instrument and will be forced to adapt the playing to fit this change and respond to the musical impulses coming from the sound processor.

Adding more live processing roles to T-EMP

After the jazz student’s workshop, we met for a rehearsal (April 2013) with the normal band configuration (Øyvind, Carl Haakon, Tone, Trond and Bernt Isak) where we aimed to incorporate more possible connections for live processing based upon our recent experiences. Trond and Bernt Isak merged their original instrument setups (using guitar as input) with the specialized live processing setup. We set up connections enabling Øyvind, Trond and Bernt Isak to process both Tone and Carl Haakon. This setup was also used during our concert in Härnösand.

Even though our new experiences and adapted instruments gave much better results than the first experiment during the Bent Sæther workshop, we still encountered some problems compared to the jazz student workshop. First of all it is a big difference in playing a specialized processing instrument with full focus compared to instruments where you can be sound producer, sound processor or both at once. The overall control intimacy decreases when your focus is spread or needs to be changed to respond to musical impulses. This has of course also to do with instrument design and lack of rehearsal, but it is a difficult mixture of roles. The other aspect that perhaps made live processing less prominent than we liked it to be, was the lack of pure acoustic sound sources. While we had between 4 and 6 acoustic sound sources to process during the jazz student workshop, we now only had Carl Haakon as pure acoustic source while Tone is sometimes acoustic, but quite often already heavily processed. It has been discussed if Tone should have two microphones; one dry for live processing and one for her regular setup, but there was no conclusion if this would be a good solution or not. During our workshop and concert with David Moss and Arnfinn Killingtveit we set up connections a bit differently; Bernt Isak had lines from Arnfinn and Carl Haakon, Øyvind had lines from Carl Haakon, David, Tone and Arnfinn, while Trond had lines from Carl Haakon, David and Arnfinn.

Experiences from practice

It is also often hard to know what kind of processing that is currently applied to your signal and therefor you will have to take care in terms of dynamics. We experienced that using short impulses is a useful “tool” to navigate the sometimes confusing sound scape that is being produced. The sound processors should also take extra care in not always filling up the entire frequency range. Equalizers are good and necessary tools to limit the different instruments and making the whole stereo image fit together instead of turning it into a big grey blob of sound. Active usage of the stereo width (i.e. panning) is also a good way of limiting the space of each instruments and contributing to a more exciting total sound

Future work: Interprocessing

During our work with extending live processing we came to an idea that was so different from “normal” live processing that we may propose the term: inter-processing. A related term intermodulation has been used by Stockhausen and others (see e.g.http://www.ears.dmu.ac.uk/spip.php?page=rubriqueLang&lang=de&id_rubrique=426), but this terms also has a specific meaning in the DSP literature where it describes the interaction between frequency components of a spectrally rich signal in a system with nonlinearities (e.g. distortion). By using the term interprocessing we wish to emphasize the use of two (or more) live signals, where features or qualities from one signal directly affects the parameters controlling the processing of the other. Ring Modulation can be used as an example of interprocessing of the most basic form.. A more complex processing form based on live convolution (see Enabling technology / Dynamic Convolution) has been developed and superficially tested during this research project, but there are still some technical issues to overcome before it can be applied properly in a musical context. The continued development of interprocessing techniques and its practical and performative application in artistic research is one of our main focus areas for future research work with T-EMP