website menu

Sound Synthesis Research in the Centre for Digital Music

Sound synthesis is the generation of sounds using algorithms, whether implemented in analogue or digital forms. It is an important application for cinema, multimedia, games and sound installations. It fits within the wider context of sound design, which is the discipline of acquiring, creating and manipulating sounds to achieve a desired effect or mood. Sound synthesis research within the Centre for Digital Music crosses several themes, including Audio Engineering and Augmented Instruments. We seek to uncover new synthesis techniques, as well as enhance existing approaches and adapt them to new applications. With a strong emphasis on performance, expression and evaluation, much of our research is focused on real world applications, empowering users and bringing sound synthesis to the forefront of sound design in the creative industries.

Some of our current and recent research projects include;

  • RTSFX (PI, Dr. Reiss, Innovate UK, 2015-16) - This project is concerned with developing and assessing a cloud-based real-time sound effects service, providing a streamlined sound synthesis-based workflow for sound designers.
  • Digital Foley Artistry (PI, Dr. McPherson, Queen Mary Innovation, 2015) - The Digital Foley project creates a rapid prototyping environment for game audio, deploying procedural audio models to embedded hardware where they can be performed using physical sensors and capturing expressive parameter trajectories from these performances.
  • Physically informed procedural audio (Researcher: Rod Selfridge, 2014 - ) – This research is concerned with developing realistic, controllable, real-time procedural audio techniques for synthesizing sound textures (e.g. wind and rain) using physical models.
  • Improved sound synthesis through perceptual evaluation (Researcher: Dave Moffat, 2014 - ) – This research seeks to advance the state of the art in sound synthesis evaluation, in order to identify the performance of synthesis techniques, both using objective and subjective measures, thus leading to deeper insights into the advantages and disadvantages of the different approaches, and guiding work towards improved techniques.

Some relevant, recent publications describing our work include;

PhD Study - interested in joining the team? We are currently accepting PhD applications.


Berker BanarTowards Composing Contemporary Classical Music using Generative Deep Learning
Marco ComunitàMachine learning applied to sound synthesis models
Max GrafPERFORM-AI (Provide Extended Realities for Musical Performance using AI)
Benjamin HayesPerceptually motivated deep learning approaches to creative sound synthesis
Yin-Jyun LuoIndustry-scale Machine Listening for Music and Audio Data
Prof Andrew McPherson
Professor of Musical Interaction
new interfaces for musical expression, augmented instruments, performance study, human-computer interaction, embedded hardware
Brendan O'ConnorSinging Voice Attribute Transformation
Prof. Joshua D Reiss
Professor of Audio Engineering
sound engineering, intelligent audio production, sound synthesis, audio effects, automatic mixing
Eleanor RowAutomatic micro-composition for professional/novice composers using generative models as creativity support tools
Shubhr SinghAudio Applications of Novel Mathematical Methods in Deep Learning
  • Dr. Andy Farnell (collaborator) – guidance on physically inspired procedural audio approaches and research directions
Return to top