The research of the group focuses on the development of signal processing and machine learning algorithms for content analysis of sounds, audio classification, sound source separation, automatic music transcription, self localization, speaker tracking, speech recognition, but includes also other areas like speech synthesis. The group has produced several state-of-the art results in the areas of audio-based music information retrieval, noise-robust automatic speech recognition, and sound event detection in realistic everyday environments.
The research is motivated by potential applications, but includes also basic research not targeting specific applications. The research is funded by different sources, including EU, Academy of Finland, Tekes, and we also have projects funded by individual companies.
More detailed description of the research areas can be found in their separate pages:
Other research units or companies interested in cooperation are encouraged to contact the group. A significant amount of the funding of the group comes from companies, so we are always interested in new research projects.
The group provides education by arranging introductory and advanced courses an audio signal processing, as well as supervising project works and theses related to audio processing.
Most of the group members are researchers working on their doctorate theses, but there are also postdoctoral senior researchers and students aiming at their M.Sc. degrees. The group cooperates actively with other national and international teams in the field. Research visits to and from our group and exchange of knowledge take place regularly with universities abroad.