The Signal Processing Group is part of the Institute of Electronics, which belongs to the Faculty of Computer Science, Electronics and Telecommunications at AGH University of Krakow, Poland. Group members perform research into various aspects of digital signal processing (DSP) focusing mainly on speech and audio signal processing for the Internet of Things (IoT), multimedia and communication applications, as well as the processing of biomedical signals and audio-video for virtual and augmented reality. Research into signal processing builds upon an intelligent integration of classical DSP techniques, statistical signal processing and machine learning.
The Signal Processing Group is led by Associate Professor Konrad Kowalczyk and it consists of over a dozen of the members of academic staff and research students at a Ph.D. level. The DSP team collaborates internationally with renown academic partners and nationally with the local high-tech industry, and has successfully completed a number of research and commercial R&D projects. The group offers fundamental and advanced taught courses on signal processing, DSP, machine learning, and programming for embedded and multimedia applications at both undergraduate and graduate levels. The Signal Processing Group regularly offers student jobs in research and R&D projects, as well as provides a breadth of opportunities for thesis work and student internships.
Several member of AGH’s Signal Processing group are heading to Sydney, Australia to attend INTERSPEECH 2026 conference to preset the recent research work on hallucinations in automatic speech recognition models. Please meet our team: Jan Jasiński, Mateusz Barański, Julitta Bartolewska, Marcin Witkowski, Konrad Kowalczyk to talk about the research performed at our lab. Wednesday 30 […]
Read More
We are proud to announce that the paper entitled “From Text Metrics to Model Internals: A Study of Whisper ASR Hallucination Detection” authored by three PhD student associated with AGH Signal Processing Group: Jan Jasiński, Mateusz Barański and Julitta Bartolewska has been nominated by ISCA (International SpeechCommunication Association) for the Best Student Papers Award at […]
Read More
We are happy to share that 2 research papers on ASR hallucinations authored by the members of the AGH Signal Processing Group were accepted for presentation at the Annual Conf. Int. Speech Communication Association (INTERSPEECH) to be held in September/October 2026 in Sydney, Australia.
Read More
We would like to invite you to the Ph.D. defense of mgr inż. Magdalena Rybicka on March the 31th 2026 which will take place in building B2, Room 4 at 12:00. The Ph.D. thesis title is “Towards discriminative speaker representations for speaker recognition and diarization”. This PhD research has been performed at AGH Univeristy and Johns […]
Read More
Thank you for visiting a poster presented by Mateusz showing the results of joint research work of Mateusz and researchers from Dolby Laboratories, Giulio Cengarle and Daniel Arteaga, entitled “Deep learning based spatial aliasing reduction in beamforming for audio capture”. The paper was presented at INTERSPEECH conference held in Rotterdam, The Netherlands, on August the […]
Read More
We are very pleased to inform that a journal article on joint separation and diarization by Magdalena Rybicka in collaboration with partners from John Hopkins University has just been published on IEEE Xplore. Title: Joint Diarization and Separation Using SepFormer With Non-Autoregressive Attractors Authors: M. Rybicka, K. Kowalczyk, T. Thebaud, N. Dehak and J. Villalba Abstract: Speaker diarization and speech separation […]
Read More
We are very pleased to announce that the joint research of our team member, mgr inż. Mateusz Guzik, and the research staff from Dolby Barcelona has been accepted for presentation at Interspeech 2025 conference, which will take place in Rotterdam, the Netherlands, from August 17 to August 21, 2025. The paper focuses on mitigating the […]
Read More
We are very pleased to announce that 4 submissions authored by members of AGH’s Signal Processing Group were accepted for presentation at the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025), which will take place in Hyderabad, India from April 06 to April 11, 2025. ICASSP paper presentation:Mateusz Barański, Jan Jasiński, Julitta Bartolewska, Stanisław Kacprzak, […]
Read More
Stanisław Kacprzak and Konrad Kowalczyk attended the IEEE Speech and Language Technology Workshop in Macau, where they presented a newly developed dataset for speaker height estimation. This dataset is detailed in the article HeightCeleb – an enrichment of the VoxCeleb dataset with dpeaker height information. A demo of the height estimation system trained using this […]
Read More
Our team member, mgr inż. Mateusz Guzik, participated in a 3-month PhD internship, during which he was working with the experts from Dolby in Barcelona, Spain. Together with the research staff from Dolby, mgr inż. Mateusz Guzik was working on deep learning approaches to spatial audio processing for beamforming applications. More information about the results […]
Read More