Impress is an automatic soundscape affect recognition system aimed toward helping sound designers to find more streamlined workflows. The system has a wide range of potential applications, including adding sound effects for films, assisting soundscape composers with creating emotional soundscape compositions trying to evoke a target mood, and enabling engineers to design mood-enabled recommendation systems for the retrieval of soundscape recordings. 

We will continue to contribute to the audio affect recognition field, implementing new tools for sound designers and artists.

Try out a version of the system on the Audio Metaphor project website.


Jianyu FanMiles Thorogood, Philippe Pasquier,

Papers & Posters

Jianyu Fan*, Miles Thorogood*, Bernhard Riecke, Philippe Pasquier. Automatic Recognition of Eventfulness and Pleasantness of Soundscape. Journal of the Audio Engineering Society – Special Issue on Intelligent Audio Processing, Oct 2016. 9 pages. (accepted).

Fan, J, Thorogood, M, and Pasquier, P. (2015). “Automatic Recognition of Eventfulness and Pleasantness of Soundscape” 10 th Audio Mostly. Thessaloniki, Greece,

Download PDF

Thorogood, M., Pasquier, P. (2013). “Impress: A Machine Learning Approach to Soundscape Affect Classification for a Music Performance Environment” Proceedings of the 13th International Conference on New Interfaces for Musical Expression (NIME). Daejeon + Seoul, Korea Republic .