You are here
The SHINE (speech-acoustic scene analysis and interpretation) unit conducted research on acoustic scene analysis and distant-speech recognition from 2008 to 2017, as part of the Center for Information Technology of Fondazione Bruno Kessler (FBK)-irst.
This research had its origins from more pioneering activities on microphone arrays and on robust speech recognition, conducted by the corresponding research units of Istituto Trentino di Cultura (ITC) - irst between 1991 and 2008.
We investigated these tasks and related techniques by addressing different application contexts, as for instance audio/video-conferencing, in-car voice interaction, and more recently distant-speech interaction in smart-homes. Most of the research activities were conducted under EC projects among which the following ones: DIMUS-II, VODIS-II, SpeechDatCar, VICO, CHIL, HIWIRE, DICIT, SCENIC, and DIRHA.
During the years, the size of the team ranged between 5 and 15 people, including researchers, technologists, and a variable number of post-doc and PhD students.
The main scientific events and communities we referred to were: ICASSP, INTERSPEECH, ASRU, EUSIPCO, HSCMA, IWAENC, WASPAA, ISMIR, LREC. You can find many of our research achievements published in the proceedings of those conferences and workshops, as well as in journal papers such as IEEE Transactions of Audio, Speech, and Language Processing, Computer Speech and Language, Speech Communication.