Till innehåll på sidan
Till KTH:s startsida Till KTH:s startsida

Spv: Kjetil Falkenberg

Project 1: Voice sketching, synthesis and musical notation of DJ scratching

Objective: Explore different representations of DJ scratching and methods for composition and notation.

Background: DJ scratching can be described in terms of isolated and named techniques, which commonly are simple synchronized gestures involving both hands. These techniques are repeated and combined in order to build up scratch performances. Although the techniques themselves are well documented, it has not been studied how they are perceived by the listener, nor if they can be reproduced vocally. Two music notation systems exist, but they have not been properly evaluated in terms of reproducing performances.

Challenges: Multidisciplinary study. Analysis and synthesis. Listening tests.

Supervisor: Kjetil Falkenberg, Claudio Panariello, Mattias Sköld

Project 2: Web audio game design for hearing impaired

Objective: Develop a music-based game for hearing impaired in WebAudio. The focus is on game design and testing. The project has elements of inclusive design and disability research.

Background: A music puzzle game for Android has been tested with hearing impaired, and we could show positive effects of using this game for training listening. However, the game was rather rudimentary, and there are many possibilities of further development and experiments, both in terms of game material, game design, code, experiment design, and applying theories of human perception.

Challenges: Real-time sound and music computing. Knowledge is welcome, but not necessary. WebAudio is fairly new and may impose some practical limitations. Hearing impairments are individual and it will be hard to find one solution that fits all.

Technologies: Mobile phones, WebAudio, sound reproduction (headphones, CI, hearing aids)

Supervisor: Kjetil Falkenberg and Hans Lindetorp

Projet 3: Design-for-all and the one-button musical instrument

Objective: Develop prototype musical instruments with only one button that are capable of rich musical output. Compare performances with traditional instruments. Suggest design-for-all guidelines for such simple interfaces.

Background: Accessible musical instruments intended for people with disabilities are typically quite simple and with little headroom for creating expressive performances. With machine learning methods, it is now within reach to quickly apply gesture or pattern recognition. Also, you will map one or several musical instruments in terms of accessibility to learn more about current regulations and implications of design for all. What can we learn from the domain of music interaction? Are current regulations, guidelines and rules appropriate for the purpose they are targeting?

Challenges: There have been other studies who approach this problem ( link ), but there is potential in machine learning that is yet to be explored.

Technologies: Programming skills, knowledge of musical instruments and playing, usability testing.

Supervisor: Kjetil Falkenberg

Project 4: Sonification against loss prevention in retail industry

Objective: Develop and test a prototype using sonification of store goods and customers for theft prevention.

Background: The retail sector in Europe calculates about 2% losses in theft and in-store damage. At the same time, store sales are losing to e-commerce. There is a need to find solutions for loss prevention (professional or opportunist shoplifters) that do not scare away customers. Sonification could be one of these solutions. Advantages of this approach is that audio systems are commonly in place, sound can be used for alerts without demanding visual attention, and goods are typically already electronically marked.

Challenges: The project is innovative without immediate works to build upon, however, sonification is a relatively well-studied area. In-situ experiment.

Technologies: Mock-up prototypes using any kind of sensing technology, or a realistic approach with e.g. RFID and situated loudspeakers. Sonification, music and audio perception. General programming skills. Focus group study, design fiction approach.

Supervisor: Kjetil Falkenberg

Project 5: Ambidexterity and interaction proficiency among musicians

Objective: To explore if musicians have a greater proficiency in two-handed operations than musically untrained persons. Especially, to see if scratch DJs who extensively use both hands for the same tasks show a higher level of ambidexterity.

Background: Scratch DJs play their instrument with both hands in a way that is unique. They are known to be able to perform synchronized, very fast and complex gestures on the crossfader and vinyl interchangeably with left and right hand. It is from a human-computer interaction point of view valuable to explore and understand this ability, and if it feasible to develop new devices for interaction that can exploit this kind of expertise.

Challenges: A typical HCI task proficiency evaluation, but with an explorative side. What and how to measure? Gesture analysis.

Technologies: Prototyping devices and suitable testing environments.

Supervisor: Kjetil Falkenberg