Shlomo E. Chazan
PhD, Research scientist
Head of Audio and NLP research group at AI Research Center.
Postdoctoral researcher at the Faculty of Engineering, Bar-Ilan University.
Research Interests:
Multi-Modal algorithms.
Deep learning for speech processing.
Single-channel enhancement.
Source localization.
Source separation.
Multi-channel beamforming with deep learning.
Deep latent models.
E2E Speech To Text (STT/ASR) models.
E2E learning approach for clustering data.
Publications:
Lior Frenkel, Shlomo E. Chazan and Jacob Goldberger, “Domain adaptation using suitable pseudo labels for speech enhancement and dereverberation”, IEEE Transactions on Audio, Speech and Language Processing. (Accepted for publication, 2024).
Amit Sofer and Shlomo E. Chazan, "“C-CLAPA: improving Text-Audio cross domain retrieval with captioning and augmentations”, IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP), 2024.
Ayal Schwartz, Elior Hadad, Sharon Gannot, Shlomo E. Chazan, "Array Configuration Mismatch in Deep DOA Estimation: Towards Robust Training",IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), 2023.
Tomer Wullach and Shlomo E. Chazan, "Optimized Tokenization for Transcribed Error Correction", Empirical Methods in Natural Language Processing (EMNLP), 2023.
Lior Frenkel, Jacob Goldberger, Shlomo E. Chazan, "Domain Adaptation for Speech Enhancement in a Large Domain Gap", INTERSPEECH, 2023.
Aviad Eisenberg, Sharon Gannot, Shlomo E. Chazan, "A two-stage speaker extraction algorithm under adverse acoustic conditions using a single-microphone", The European Signal Processing Conference (EUSIPCO), 2023.
Tomer Wullach and Shlomo E. Chazan, "Don’t Be So Sure! Boosting ASR Decoding via Confidence Relaxation", Advancement of Artificial Intelligence (AAAI), 2023.
Aviad Eisenberg, Sharon Gannot, and Shlomo E. Chazan, "Single microphone speaker extraction using unified time-frequency Siamese-Unet.", The European Signal Processing Conference (EUSIPCO), 2022.
Shlomo E. Chazan, Lior Wolf, Eliya Nachmani and Yossi Adi, ”Single Channel VoiceSeparation for Unknown Number of Speakers Under Reverberant and Noisy Settings,”IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP),2021.
Shlomo E. Chazan, Jacob Goldberger and Sharon Gannot, ”Speech Enhancement withMixture of Deep Experts with Clean Clustering Pre-Training,”, IEEE International Con-ference on Acoustic, Speech and Signal Processing (ICASSP), 2021.
Hodaya Hammer, Shlomo E. Chazan, Jacob Goldberger, and Sharon Gannot. ”Dynamically localizing multiple speakers based on the time-frequency domain.” EURASIP Journal on Audio, Speech, and Music Processing 2021, no. 1 (2021): 1-10
Yochai Yemini, Shlomo E. Chazan, Jacob Goldberger and Sharon Gannot, “A composite DNN Architecture for Speech Enhancement”,IEEEInternational Conference on Acoustic, Speech and Signal Processing (ICASSP), 2020.
Yaniv Opochinsky, Shlomo E. Chazan, Sharon Gannot and Jacob Goldberger, “k-Autoencoders Deep Clustering”,IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP), 2020.
Shlomo E. Chazan, Sharon Gannot and Jacob Goldberger, “Deep clustering based on a mixture of autoencoders”, IEEE Machine Learning for Signal Processing Workshop (MLSP), 2019.
Shlomo E. Chazan, Hodaya Hammer, Gershon Hazan, Jacob Goldberger and Sharon Gannot, “Multi-microphone speaker separation based on deep DOA estimation”,The European Signal Processing Conference (EUSIPCO), 2019.
Shlomo E. Chazan, Sharon Gannot and Jacob Goldberger, “Attention-based neural network for joint diarization and speaker extraction”, Int.Workshop on Acoustic Signal Enhancement (IWAENC), 2018.
Shlomo E. Chazan, Jacob Goldberger and Sharon Gannot, “LCMV beamformer with DNN-based multichannel concurrent speakers detector”,The European Signal Processing Conference (EUSIPCO), 2018.
Ori Ernst, Shlomo E. Chazan, Sharon Gannot and Jacob Goldberger, “Speech dereverberation using fully convolutional networks”,The European Signal Processing Conference (EUSIPCO), 2018.
Shlomo E. Chazan, Sharon Gannot and Jacob Goldberger, “Training strategies for deep latent models and applications to speech presence probability estimation”, Int. Conference on Latent Variable Analysisand Signal Separation (LVA/ICA), 2018.
Hodaya Hammer, Gilad Rath, Shlomo E. Chazan, Jacob Goldberger and Sharon Gannot, “Speech enhancement with deep neural networks using MoG based labels”,Int. Conference on the Science of Electrical Engineering (ICSEE), 2018.
Shlomo E. Chazan, Jacob Goldberger and Sharon Gannot, “DNN-based concurrent speaker detector and its application to speaker extraction with LCMV beamforming”, IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP), 2018.
Shlomo E. Chazan, Jacob Goldberger and Sharon Gannot, “Deep recurrent mixture of experts for speech enhancement”,IEEE Workshop on Applications of Signal Processing to Audio and Acoustics(WASPAA), 2017.
Dany Cherkassky, Shlomo E. Chazan, Jacob Goldberger and Sharon Gannot, “Successive relative transfer function identification using single microphone speech enhancement”, The European Signal Processing Conference (EUSIPCO), 2017.
Ariel Malek, Shlomo E. Chazan, Ilan Malka, Vladimir Tourbabin, Jacob Goldberger, Eli Tzirkel-Hancock and Sharon Gannot, “Speaker extraction using LCMV beamformer with DNN-based SPP and RTF identification scheme”,The European Signal Processing Conference (EUSIPCO), 2017.
Shlomo E. Chazan, Sharon Gannot and Jacob Goldberger, “A pre-training approach for deep neural network with application to speech enhancement”, Int. Workshop on Acoustic Signal Enhancement (IWAENC), 2016. Best student paper award.
Shlomo E. Chazan, Jacob Goldberger and Sharon Gannot, ”A hybrid approach for speech enhancement using MoG model and neural network phoneme classifier”, IEEE Transactions on Audio, Speech and Language Processing, vol. 24(12), pp. 2516-2530, 2016.