|
Publications
Theses
Journal
“CL-MASR: A Continual Learning Benchmark for Multilingual ASR”, Submitted to IEEE/ACM Transactions on Audio, Speech, and Language Processing (TASL), Luca Della Libera, Pooneh Mousavi, Salah Zaiem, Cem Subakan, Mirco Ravanelli arxiv_link
“Exploring Self-Attention Mechanisms for Speech Separation”, IEEE/ACM Transactions on Audio, Speech, and Language Processing (TASL), Cem Subakan, Mirco Ravanelli, Samuele Cornell, François Grondin, Mirko Bronzi arxiv_link
“Learning Representations for New Sound Classes With Continual Self-Supervised Learning”, Zhepei Wang, Cem Subakan, Xilin Jiang, Junkai Wu, Efthymios Tzinis, Mirco Ravanelli, Paris Smaragdis, arxiv_link
“Probabilistic Sequence Clustering with Spectral Learning”, Elsevier Digital Signal Processing (DSP), Subakan, Y.C.; Kurt, B. ; Cemgil, A.T; Sankur, B. elsevier_link
Conference
“CryCeleb: A Speaker Verification Dataset Based on Infant Cry Sounds”, ICASSP 2024, David Budaghyan, Charles C. Onu, Arsenii Gorin, Cem Subakan, Doina Precup, arxiv_link
“Resource-efficient separation transformer”, ICASSP 2024, Luca Della Libera, Cem Subakan, Mirco Ravanelli, Samuele Cornell, Frédéric Lepoutre, François Grondin, arxiv_link
“Unsupervised Improvement of Audio-Text Cross-Modal Representations”, WASPAA 2023, Zhepei Wang, Cem Subakan, Krishna Subramani, Junkai Wu, Tiago Tavares, Fabio Ayres, Paris Smaragdis, arxiv_link
“CommonAccent: Exploring Large Acoustic Pretrained Models for Accent Classification Based on Common Voice”, Interspeech 2023, Juan Zuluaga-Gomez, Sara Ahmed, Danielius Visockas, Cem Subakan, arxiv_link
“REAL-M: Towards Speech Separation on Real Mixtures”, ICASSP 2022, C. Subakan, M. Ravanelli, S. Cornell, F. Grondin, arxiv_link
“Attention is all you need in speech separation”, ICASSP 2021, C. Subakan, M. Ravanelli, S. Cornell, M. Bronzi, J Zhong arxiv_link
“On the Effectiveness of Two-Step Learning for Latent-Variable Models”, MLSP 2020, C. Subakan, M. Gasse, L. Charlin
“A generative modeling approach for interpreting population-level variability in brain structure”, MICCAI 2020, R. Liu, C. Subakan, A.H. Balwani, J.D. Whitesell, J.A. Harris, S. Koyejo, E. Dyer pdf
“Two-Step Sound Source Separation: Training On Learned Latent Targets”, ICASSP 2020, E. Tzinis, S. Venkataramani, Z. Wang, C. Subakan, P. Smaragdis pdf
“Continual Learning of New Sound Classes Using Generative Replay”, WASPAA 2019, Z. Wang, C. Subakan, E. Tzinis, P. Smaragdis, L. Charlin pdf
“Neural Network Alternatives to Convolutive Models for Source Separation”, (Won the best student paper award) IEEE Workshop on Machine Learning for Signal Processing (MLSP) 2017, S. Venkataramani; Y.C. Subakan; P. Smaragdis arxiv_link
“Method of Moments for Left-to-Right Hidden Markov Models”, IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), 2015, Y. C. Subakan; J. Traa; P. Smaragdis; D. Hsu pdf
“Spectral Learning of Mixture of Hidden Markov Models”, Neural Information Processing Systems (NIPS), 2014, Y. C. Subakan; J. Traa; P. Smaragdis pdf
“Stereo Matching with Partial Information”, SPIE 3D Image Processing (3DIP) and Applications 2013, Subakan, Y.C.; Gurol, O.C.; Dikici, C." pdf
“Model Based Probabilistic Piecewise Curve Approximation”, European Signal Processing Conference (EUSIPCO) 2012, Subakan, Y.C.; Sankur, B. pdf
“Probabilistic Latent Tensor Factorization Framework for Audio Modeling”, in Proc. IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), 2011, Cemgil, A.T; Simsekli, U.; Subakan, Y.C. pdf
Conference (In Turkish)
“Spectral Learning of Mixtures of Hidden Markov Models”, IEEE 21th Signal Processing and Communications Applications Conference (SIU), 2013, Y. C. Subakan; O. Celiktutan; A. T. Cemgil; B. Sankur (in Turkish) pdf
“Model Based Bayesian Curve Compression”, (Nominated for best paper award), IEEE 20th Signal Processing and Communications Applications Conference (SIU), 2012, Subakan, Y.C.; Sankur, B.; Akgul, C.B. (in Turkish) pdf
“A latent tensor factorization framework for non-negative convolutive models”, IEEE 19th Signal Processing and Communications Applications Conference (SIU), 2011, Simsekli, U.; Subakan, Y.C.; Cemgil, A.T. (in Turkish) pdf
“An automatic maneuver assessment system for radio controlled acrobatic model helicopter competitions”, IEEE 19th Signal Processing and Communications Applications Conference (SIU), 2011, Subakan, Y.C.; Agcaoglu, O.; Sankur, B. (in Turkish) pdf
Workshop
“Self-supervised learning for infant cry analysis”, ICASSP 2023 Workshop on Self-Supervised Learning for Speech, Audio and Beyond, Arsenii Gorin, Cem Subakan, Sajjad Abdoli, Junhao Wang, Samantha Latremouille, Charles Onu, arxiv_link
‘‘Towards Lifelong Learning via Maximum Entropy Generative Replay’’, ICML 2019 Workshop on Continual Learning, Y.C. Subakan, M. Caccia, T. Lesort, L. Charlin
" Spectral Learning for Mixture of Markov Models’’, Spectral Learning Workshop, Neural Information Processing Systems (NIPS) 2013, Subakan, Y.C.; Kurt, B. ; Cemgil, A.T; Sankur, B. pdf, supplemental, poster
" Spectral Learning of Infinite Mixtures of Hidden Markov Models for Human Action Recognition’’, Spectral Learning Workshop, Neural Information Processing Systems (NIPS) 2012, Subakan, Y.C.; Celiktutan O. ; Cemgil, A.T; Sankur, B. pdf
Preprints
‘‘A Latent Template Network for Discriminative Source Separation Using Attention Mechanism’’, 2018, Y.C. Subakan, M. Caccia, L. Charlin
“A Dictionary Learning Approach for Factorial Gaussian Models”, 2015, Y. C. Subakan; J. Traa; P. Smaragdis; N. Stein arxiv_link, supplemental_mat
“Spectral Learning of Hidden Markov Models with Group Persistence”, 2015, Y. C. Subakan; J. Traa; P. Smaragdis pdf, supplemental_mat
“Efficient Learning for Time Series Models by Non-Negative Matrix Factorization”, 2014, Y. C. Subakan; J. Traa; P. Smaragdis pdf, supplemental_mat
|