Breaking down barriers in professional music production

Research by Dr Ryan Stables and Dr Jason Hockman has led to the development of intelligent audio software which makes the technical and creative process of audio mixing easier for both professionals and hobbyists. This suite of tools has been used by professional music studios to create commercial recordings and has been downloaded by 15,000 music-creators in 20 countries.

The research team also developed this technology further, expanding the concept into a commercial spinout which provides sophisticated production tools to businesses and creators.

Research summary

The research team sought to lower the barriers of entry associated with audio production and engineering, which usually requires extensive experience or investment to achieve the desired results.

Focusing initially on improving the experience for amateur musicians and podcasters, the team utilised their research into machine learning and automation to develop systems that streamline the production process.

These systems reduce the complexity of sound engineering, using machine learning to mix music automatically, simplify the control of audio effects and take care of several other difficult tasks.

This work resulted in the development of a bespoke Digital Audio Workstation (DAW) called Faders. DAWs are tools all sound designers use to process audio, but Faders is embedded with intelligent quality of life tools, making audio production faster and easier.

Research background

Dr Stables and the team developed several tools to lower the barriers to professional sound production, each tackling a different technical area.

Semantic Audio Feature Extraction (SAFE) plugins allow engineers to easily apply audio effects to music samples by using related descriptions i.e. warm, light, dramatic.

Faders is a DAW which utilises multiple intelligent systems that assist creatives in mixing, editing regions, labelling recordings, creating content and collaborating in real time with multiple partners. The suite allows both professional and aspiring creators to work more efficiently.

Research outcomes

The software systems, SAFE plugins and Faders, developed by Stables and his team promoted a change in audio engineering practice by providing enhanced functionality to audio content creators with a wide range of backgrounds and levels of experience.

SAFE plugins have been installed in several professional recording studios around the world and have been used on high grossing commercial recordings. The programmes allow creators to focus on the creative process while offsetting some of the technical tasks, in some instances allowing work to be completed in half the time.

This success has seen the toolset grow into a spinout company called Semantic Audio Labs, which provides access to innovative audio tools through the web.

REF 2021

Birmingham City University's hub for everything related to the Research Excellence Framework.

REF 2021


Ryan Stables DMT staff

Ryan Stables

Ryan Stables is an Associate Professor of Audio Processing, based in the Digital Media Technology (DMT) Lab. 

He currently teaches in the areas of Digital Audio Effects, Digital Signal Processing and Audio Software Development.

Ryan currently runs Semantic Audio Labs, a university spin-out company that builds intelligent audio production tools. He has received international news coverage for his work on intelligent music production and cancer diagnostics via data sonification.


Jason Hockman

Jason Hockman

Associate Professor

Jason Hockman is Associate Professor of Audio Engineering at Birmingham City University. He is a member of the Digital Media Technology Laboratory (DMT Lab), in which he leads the Sound and Music Analysis Group (SoMA) for computational analysis of sound and music and digital audio processing.    

Jason's expertise is in music informatics, machine listening and computational musicology—topics in which he lectures and has published extensively. His research has focused on a variety of aspects related to computational rhythm and metre detection, music transcription, and content-based audio effects. Jason has received a Bachelor’s degree in Sociology from Cornell University (USA, 2000), a Masters in Music Technology from New York University (USA, 2007), and a PhD in Music Research from McGill University (Canada, 2014).