Audio and Speech Processing Lab focuses on applied research problems and aims at developing new software tools applicable in real-life scenarios.
Our main research topics are:
- Audio and Speech Signal Processing
- Computational Musicology
- Music Information Retrieval
- Automatic Speech Recognition
- Speaker Verification
- Natural Language Processing
Selected projects lead in our lab:
- Automatic melodic analysis of Turkish makam music, TÜBİTAK ARDEB 1001. 2012-2014.
- Automatic Transcription of Turkish music, TÜBİTAK ARDEB 1001. 2007-2010.
- Speech and Speaker Recognition for Mobile Devices, TUBITAK ARDEB 3001. 2014-2016.
Selected projects we took part:
- CompMusic, EU(ERC-267583) [PI: Prof. Xavier Serra], Computational models for the discovery of the World’s Music. 2011 – 2017
- Automation of corporate support services using chat-bot, supported by TUBITAK TEYDEB, the project is carried out at Sestek Conversational Solutions Technology Company. 2018-2021
- Analysis and optimization of user experience for human-machine interaction, supported by TUBITAK TEYDEB, the project is carried out at Sestek Conversational Solutions Technology Company. 2016 –2017
Selected publications of our lab:
- B Bozkurt, I Germanakis, Y Stylianou, 2018, A study of time-frequency features for CNN-based automatic heart sound classification for pathology detection, Computers in Biology and Medicine 100, 132-143.
- T Drugman, B Bozkurt, T Dutoit, 2012, A comparative study of glottal source estimation techniques, Computer Speech & Language 26 (1), 20-34.
- AC Gedik, B Bozkurt, 2010, Pitch-frequency histogram-based music information retrieval for Turkish music, Signal Processing 90 (4), 1049-1063
- A Holzapfel, Y Stylianou, AC Gedik, B Bozkurt, 2010, Three dimensions of pitched instrument onset detection, IEEE Transactions on Audio, Speech, and Language Processing 18 (6), 1517-1527.
- B Bozkurt, R Ayangil, A Holzapfel, 2009, Computational analysis of turkish makam music: Review of state-of-the-art and challenges, Journal of New Music Research 43 (1), 3-23.
- B Bozkurt, L Couvreur, T Dutoit, 2007, Chirp group delay analysis of speech signals, Speech communication 49 (3), 159-176.
- O Buyuk, 2020, Context-dependent sequence-to-sequence Turkish spelling correction, ACM Transactions on Asian and Low-Resource Language Information Processing 19 (4), article 56, 1-16.
- C. Demiroglu, O. Buyuk, A. Khodabakhsh, R. Maia, 2017, Post-processing synthetic speech with a complex cepstrum vocoder for spoofing phase-based synthetic speech detectors, IEEE Journal of Selected Topics in Signal Processing 11 (4), 671-683.
- K Korucu, O Kaplan, O Buyuk, K Gullu, 2016, An investigation of usability of sound recognition for source separation of packaging waste in reverse vending machines, Waste Management 56, 46-52.
- O Buyuk, 2016, Sentence-HMM state based i-vector/PLDA modeling for improved performance in text dependent single utterance speaker verification, IET Signal Processing 10 (8), 918-923.