I am a researcher and Phd student at Laboratory of Signal Processing (Tampere University of Technology). My primary research interest is in content analysis of audio signals, and recent years I have been focusing especially on environmental audio. During the years, I have worked in numerous academic and industrial projects related to the content analysis of audio signals. On my free time, I like to play basketball and develop various personal coding projects.
My current research topics include automatic sound event detection, automatic audio context recognition and auditory scene synthesis. I have worked on various research topics related on audio signal processing, audio classification, and audio content analysis since 2000 as a member of the Audio Research Group. In my master studies, I studied musical genre classification and other music classification problems. Later, I concentrated a few year on musical instrument classification in multi-source situation. More recently, I have studied topics which falls under the Computational Auditory Scene Analysis, e.g. sound event detection in multi-source situations.
DCASE Challenges and Workshops
Audio research group from TUT has been part of organizing teams of the last two most recent Detection and Classification of Acoustic Scenes and Events (DCASE) evaluation campaigns and workshops: DCASE2016, DCASE2017, and DCASE2018. The DCASE aims to collect researchers interested on environmental sound classification and detection together under one community. This community offers a platform for discussion of the different perspectives and approaches, from algorithm development to practical applications and their commercial value. In DCASE2016 and DCASE2017, I have acted as one of the task coordinators for the evaluation campaign (acoustic scene classification and sound event detection tasks), and publication chair for the workshop. In DCASE208, I acted as the task coordinator for the acoustic scene classification task. In the role of task coordinator, I have designed evaluation task setups and evaluation metrics, prepared the public datasets and implemented the baseline systems. This work has resulted many public datasets, reference systems, and evaluation metric implementations.
I have worked in numerous academic and industrial projects related to the content analysis of audio signals. In these projects, I have developed many prototype systems and research result demonstrators. Some of these are publicly available:
During the years, I have supervised some student projects and a few master thesis works at TUT.