Deciphering Room Acoustics
Who: Nils Peters - ICSI and CNMAT at UCB
What: Room Identification using Acoustic Features
When: Friday December 7th at 1:15PM
Where: CCRMA Seminar Room
Why: Because reverb is interesting
While the CCRMA Seminar room is small, the discussions we have reverberate throughout the world. Come to CCRMA and be part of the fun!
Name That Room: Room Identification Using Acoustic Features in a Recording
This paper presents a system for identifying the room in an audio or video recording through the analysis of acoustical properties. The room identification system was tested using a corpus of 13440 reverberant audio samples. With no common content between the training and testing data, an accuracy of 61% for musical signals and 85% for speech signals was achieved. This approach could be applied in a variety of scenarios where knowledge about the acoustical environment is desired, such as location estimation, music recommendation, or emergency response systems.
Nils Peters is a postdoctoral fellow at the International Computer Science Institute (ICSI) and the Center for New Music and Audio Technologies (CNMAT) at UC Berkeley, California. He holds a MSc degree in Electrical and Audio Engineering from the University of Technology in Graz, Austria, in 2005 and a PhD in Music Technology from McGill University in Montreal, Canada, in 2011. Besides being a core developer of the media processing library Jamoma, he works on real-time algorithms for sound-field analysis using microphone arrays. Nils also organizes UC Berkeley’s lecture series on spatial audio technology and room acoustics.