Date: 14 Oct 2014
This month’s lecture will showcase cutting edge research from City University’s Music Informatics Research group and King’s College London’s Centre for Telecommunications Research. The evening will include a drinks reception a selection of technology-based creative works from the Music Department at City.
Music Informatics Research in the Department of Computer Science
The Music Informatics Research Group in the Department of Computer Science works since 2005 on analysing music as audio and symbolic data (scores, MIDI). At City we bring together expertise in machine learning, signal processing, computer science and musicology to develop intelligent music analysis and processing methods. We work on challenges such as audio transcription, music audio similarity, music voice separation, chord recognition and melody models for generation and classification of music. Our work focuses on analysing and recognising musical structure and we are interested in particular in the integration of audio and symbolic music representations and processing. We use our methods in interdisciplinary applications such as music education software and a game interface for data collection, as well as large scale processing of music audio and scores for musicology and music retrieval.
- Automatic Music Transcription: Methods and Applications
- Big Data for Musicology and Music Retrieval
- Music Audio Similarity Models and a Game with a Purpose
- Music Language Models
Composition Research in the Department of Music
At City we recognise that the ways in which composers create and share work are shifting and changing. Traditionally delineated boundaries between the fields of scored concert music, studio composition, and media composition are increasingly dissolving to form a broad and fluid landscape for contemporary composers. For the music department at City, this broad field of contemporary composition encompasses notated and digital music, sound arts, improvisation, interdisciplinary practices and numerous points of intersection between these areas. At present staff and students are engaged in practice-led research in instrumental composition, live electronic performance, multichannel studio composition, interdisciplinary and collaborative research (notably in music and dance, and music and film), and sound installation theory and practice. Composition in this broad sense forms a critical strand of research in music at City alongside the department’s other strengths in musicology, ethnomusicology and performance research.
Audio Lab, Centre for Telecommunications Research, King’s College London
The research of the Audio Lab at King’s College London is centred on multichannel systems for perceptual sound field synthesis and reproduction. The field of spatial sound has so far been mainly geared towards creating special effects and providing a pleasing listening experience, rather than rooted in solid engineering or science. Notable exceptions include ambisonics and WFS, which unfortunately haven’t penetrated the market yet. At King’s, we established a scientific framework for the analysis and design of multichannel systems based on concise modelling of underlying psychoacoustic phenomena. That framework enabled the development of a new multichannel audio technology which improves over state-of-the-art systems in terms of accuracy and stability of the auditory perspective. We also developed a super-real-time software implementation for virtual reality applications, based on further psychoacoustic approximation, as well as a new class of underlying higher-order microphones.
- Perceptual Sound Field Recording, Reproduction, and Synthesis
- Efficient Synthesis of Room Acoustics Via Scattering Delay Network
- A New Class of Higher Order Differential Microphones
- A Computational Model for the Prediction of Localisation Uncertainty