Our Digital Life Podcast: A series by IEEE-SPS

In this episode, Prof. Hung-yi Lee, Professor at National Taiwan University, interviews Dr. Jinyu Li, Partner Applied Science Manager at Microsoft. Their conversation traces the evolution of speech translation from traditional cascaded pipelines to today's end-to-end, large language model (LLM)-powered systems. Jinyu Li Dr. Jinyu Li is Partner Applied Science Manager at Microsoft, where he leads a science team advancing speech, translation, and language technologies. He is an IEEE Fellow and an ISCA Fellow for contributions to deep-learning-based speech technology innovation, adoption, and commercialization. He has served on the IEEE Speech and Language Processing Technical Committee and as Vice Chair starting in 2026. He previously served as Associate Editor of IEEE/ACM Transactions on Audio, Speech, and Language Processing and is a Distinguished Industry Speaker for the IEEE Signal Processing Society. He has also been awarded the IEEE SPS Best Paper Award (2025), the APSIPA Industrial Distinguished Leader Award (2021), and the APSIPA Sadaoki Furui Prize Paper Award (2023). In this episode, Dr. Li discusses the evolution of AI-powered speech translation, from conventional cascaded systems to modern end-to-end architectures, while highlighting the role of large language models in enabling real-time multilingual communication. He also explores the latest applications of speech translation and examines the technical challenges that remain.

In this episode, Nantheera Anantrasirichai (Pui), Professor of Visual Computing at the University of Bristol, interviews Professor Alin Achim, Chair of Computational Imaging at the University of Bristol. Their conversation focuses on synthetic aperture radar (SAR), interferometric synthetic aperture radar (InSAR), and the role of computational imaging, signal processing, and machine learning in earth observation. Alin Achim Professor Alin Achim is Chair of Computational Imaging at the University of Bristol. He received his B.Sc. and M.Sc. degrees in electrical engineering from “Politehnica” University of Bucharest, Romania, and his Ph.D. in Biomedical Engineering from the University of Patras, Greece. He subsequently held an ERCIM Postdoctoral Fellowship at ISTI-CNR in Pisa, Italy, and INRIA Sophia Antipolis, France, before joining the University of Bristol in 2004. Since August 2018, he has held the Chair of Computational Imaging. His research interests include statistical signal, image, and video processing and machine learning, with applications in biomedical imaging and earth observation. He has coauthored more than 200 scientific publications, including 66 journal articles, and has contributed to several IEEE technical committees and editorial boards. In this episode, Professor Achim highlights computational imaging, statistical signal and image processing, and machine learning in the context of earth observation.

In this episode, Marco Huber speaks with Marija Ivanovska, Assistant at the Faculty of Electrical Engineering, University of Ljubljana, Slovenia. They discuss face morphing attacks and their impact on biometric identity systems, exploring how morphed images can deceive face recognition technologies used in passports and ID documents. The conversation also examines emerging approaches to detecting these attacks, including foundation models, multimodal AI, synthetic data, and other advances in trustworthy biometrics. Marija Ivanovska Marija Ivanovska is an Assistant at the Faculty of Electrical Engineering, University of Ljubljana, Slovenia, and a member of the Laboratory for Machine Intelligence. Her research focuses on biometric security, computer vision, and trustworthy artificial intelligence, with particular emphasis on face-based attacks on biometric systems and the use of foundation models and multimodal large language models for robust biometric security. She has held international research appointments as a Visiting Scholar at Johns Hopkins University and as a Visiting Researcher at Queensland University of Technology, fostering collaborations in biometrics and trustworthy AI. She has authored numerous publications in peer-reviewed journals and conferences in computer vision, biometrics, and information forensics. She is actively involved in the international research community through editorial, organizational, and reviewing roles. Marija is a member of the IEEE Information Forensics and Security Technical Committee and the IAPR Technical Committee on Biometrics and serves on the Webinar Committee of the IEEE Biometrics Council. She is passionate about advancing trustworthy biometric technologies and encouraging collaboration in AI and biometrics to help build more secure, reliable, and resilient digital identity systems.

In this episode of the IEEE Signal Processing Society podcast, Dr. Yan Qiao, an Associate Professor at the School of Computer Science and Information Engineering, Hefei University of Technology, interviews Professor Meng Li from Hefei University of Technology, whose research focuses on applied cryptography, secure and trustworthy data sharing, blockchain, privacy preservation, and Trusted Execution Environments (TEE). Professor Meng Li Professor Meng Li is a Professor at the School of Computer Science and Information Engineering, Hefei University of Technology (HFUT), China. He earned his Ph.D. in Computer Science and Technology from the Beijing Institute of Technology and has held several international research appointments, including postdoctoral positions in Italy supported by the ERCIM "Alain Bensoussan" Fellowship Programme and research collaborations at the University of Waterloo, Wilfrid Laurier University, and the University of Padua. His research focuses on secure and trustworthy data sharing, privacy preservation in the Internet of Vehicles (IoV), applied cryptography, blockchain, and Trusted Execution Environments (TEE). He is a Senior Member of IEEE, CCF, CACR, CIE, and CIC, and has been recognized with the 2024 IEEE HITC Award for Excellence (Early Career Researcher), the 2025 IEEE TCSVC Rising Star Award, and selection among the IEEE Computer Society Computing's Top 30 Early Career Professionals for 2025. In this episode, Professor Li explains why secure and trustworthy data sharing has become a cornerstone of today's digital world. He discusses how technologies such as cryptography, blockchain, and privacy-preserving techniques help enable secure data sharing while protecting user privacy and explores the role of emerging technologies in addressing modern data-sharing challenges and advancing humanitarian and societal applications.

In this episode of the IEEE Signal Processing Society podcast, Dr. Rogério Augusto Bordini, a Post-doctoral Researcher and Science Journalist at the Artificial Intelligence Lab., Recod.ai, University of Campinas (Unicamp), interviews Dr. Anderson Rocha, Full Professor at the University of Campinas (Unicamp) specializing in Artificial Intelligence, Digital Forensics, and Reasoning for Complex Data. Their conversation explores how modern signal-processing techniques have been explored in various social sectors. Dr. Anderson Rocha Professor Anderson Rocha, Former Director of Unicamp's Institute of Computing and two-time Chair of the IEEE Information Forensics and Security Technical Committee, was named an IEEE Fellow in 2023, an IEEE SPS Distinguished Lecturer in 2025, and an IEEE Biometrics Council Distinguished Lecturer also in 2025. Closing a remarkable year, he was awarded the prestigious Zeferino Vaz Prize—Unicamp's highest recognition. Recognized as one of the world's top scientists by Stanford, PLOS ONE, and Research.com, he holds fellowships from Microsoft and Google and co-founded the Recod.ai AI Lab at Unicamp over 16 years ago. In this episode, Dr. Rocha discusses applications of signal processing spanning digital forensics, wearable sensors, deepfake detection, and misinformation mitigation, while highlighting core algorithms, real-world healthcare applications, and emerging AI-driven forensic tools, and also provides insights into his research group's key differentiator—a humanistic, expert-in-the-loop approach to solution design.

In this episode of the IEEE Signal Processing Society Podcast, Hemang Chawla, Solutions Lead at Scantrust, speaks with Justin Picard, Co-founder and CTO of Scantrust. Their conversation explores how modern signal processing, printing physics, and machine learning are being combined to combat the global problem of product counterfeiting through secure QR codes and copy detection technology. Dr. Justin Picard Dr. Justin Picard is the Co-founder and Chief Technology Officer of Scantrust, a company specializing in product authentication and traceability solutions. Originally from Canada and now based in Switzerland, Dr. Picard completed his Ph.D. in artificial intelligence before moving into digital watermarking and image security. After working in research and development roles across North America and Europe, Dr. Picard co-founded Scantrust to develop smartphone-based authentication systems that empower consumers and brands to verify product authenticity in real time. In this episode, Dr. Picard discusses the trillion-dollar global impact of counterfeiting, which now affects not only luxury goods but also everyday products such as food, industrial components, health supplements, and consumer goods—an issue intensified by e-commerce and global supply chains. He explains that traditional anti-counterfeiting methods, including holograms, UV inks, and forensic testing, struggle to scale in today’s digital marketplace because they rely on specialized equipment or human inspection.

Functional Brain Imaging: Signals, Imaging, and Graphs In this episode of the IEEE Signal Processing Society Podcast, Professor Borbála Hunyadi from the Mental Health and Neuroscience Research Institute, Maastricht University, The Netherlands interviews Dr. Dimitri Van De Ville, Full Professor at the École Polytechnique Fédérale de Lausanne (EPFL) and the University of Geneva, Switzerland. Their conversation explores how modern neuroimaging modalities, combined with advanced signal processing and computational methods, are transforming our understanding of brain function in health and disorder. Dr. Dimitri Van De Ville Dr. Dimitri Van De Ville received his M.S. and Ph.D. degrees from Ghent University, Belgium, in 1998 and 2002, respectively. He was a postdoctoral fellow at EPFL before leading the Signal Processing Unit at the University Hospital of Geneva as part of the CIBM Center for Biomedical Imaging. Since 2024, he has been a Full Professor at EPFL’s Neuro-X Institute with a joint appointment at the University of Geneva. His interdisciplinary research focuses on computational neuroimaging, wavelets, sparsity, and graph signal processing, applied to MRI and M/EEG data. In this episode, he discusses current and emerging neuroimaging modalities such as intracranial recordings, fMRI, fNIRS, M/EEG, and functional ultrasound (fUS). He highlights how signal processing plays a vital role in data formation, preprocessing, and analysis, enabling researchers to extract meaningful information about brain activity. The discussion also touches on innovations such as independent component analysis, connectomics, and the growing influence of AI and deep learning in neuroimaging. Dr. Van De Ville concludes by reflecting on the field’s future—emphasizing multimodal integration, brain–body connectivity, and targeted neuromodulation as key directions for advancing both neuroscience research and clinical applications.

In this episode of the IEEE Signal Processing Society podcast, Felicia Lim, a staff software engineer at Google, where she works on audio signal processing and machine learning, interviews Dr. Ivan Tashev, Partner Software Architect at Microsoft Research (MSR) – Redmond USA, where he leads the Audio and Acoustics Research Group. Their conversation explores the rapid development of novel algorithms in AI and their impact on the audio processing domain. Dr. Ivan Tashev Dr. Ivan Tashev is a Partner Software Architect at MSR in Redmond, WA, USA, where he leads the Audio and Acoustics Research Group and also coordinates the Brain-Computer Interfaces project. He is an Affiliate Professor in the Department of Electrical and Computer Engineering at the University of Washington in Seattle, USA, and an Honorary Professor at the Technical University of Sofia, Bulgaria. He is also an IEEE Fellow and a member of the Audio Engineering Society (AES) and the Acoustical Society of America (ASA). In this episode, Dr. Tashev discusses the unique challenges of audio signal processing as a specialized domain, examining why traditional statistical methods have limitations and how machine learning and AI approaches offer new solutions. He also talks about the future trajectory of machine learning and AI in transforming audio signal processing capabilities.

In this episode of the IEEE Signal Processing Society podcast, Dr. Lav Varshney, Associate Professor of Electrical and Computer Engineering at the University of Illinois Urbana-Champaign interviews Dr. Kush Varshney, an IBM Fellow and globally recognized expert in trustworthy machine learning. Their conversation explores the multifaceted landscape of trustworthy AI. Kush Varshney Kush R. Varshney is an IBM Fellow at IBM Research and a leading authority on trustworthy AI. His work focuses on making AI systems not only accurate but also fair, robust, explainable, transparent , inclusive, and beneficial. He is the author of a book entitled “Trustworthy Machine Learning” and creator of widely used toolkits like AI Fairness 360 and AI Explainability 360. In this episode, Dr. Varshney outlines the core principles of trustworthy AI and distinguishes it from related concepts such as AI ethics, AI safety, and responsible AI. He shares how signal processing techniques—like Boolean compressed sensing and continued fraction representations, and short-time Fourier transforms—inform his approach. The conversation covers the societal impact of AI, the shift toward generative and agentic models, the importance of governance and policy, and new research directions aimed at building more empowering and accountable AI systems.

In this episode of the IEEE Signal Processing Society podcast, Nir Shlezinger from Ben-Gurion University and Yonina C. Eldar from the Weizmann Institute of Science discuss the design of machine learning systems that are inherently efficient. Nir Shlezinger and Yonina C. Eldar Nir Shlezinger is an Assistant Professor in the School of Electrical and Computer Engineering at Ben-Gurion University of the Negev, Israel. His research spans signal processing, machine learning, and communications. He has been recognized with several prestigious awards, including the IEEE Communications Society Fred W. Ellersick Prize and the 2024 Krill Award. Yonina C. Eldar is a Professor at the Weizmann Institute of Science, where she heads the Center for Biomedical Engineering and Signal Processing. She is also a member of the Israel Academy of Sciences and Humanities and an IEEE Fellow. In this episode, Dr. Shlezinger and Dr. Eldar engage in a rich discussion on model-based deep learning—an approach that combines classical signal processing principles with modern data-driven techniques. This framework promotes efficiency not only through computational improvements, but by designing learning algorithms that naturally align with physical models and mathematical structures. They explore the key principles behind this methodology, its practical advantages, and its growing impact across a range of signal processing applications.

In this episode of the IEEE Signal Processing Society podcast, Dr. Sanjeev Khudanpur, Director of the Center for Language and Speech Processing, Johns Hopkins University interviews Associate Prof. Shinji Watanabe, Language Technologies Institute, Carnegie Mellon University. They talk about the latest research and innovations in speech recognition technologies and their impact across various industries. Shinji Watanabe Shinji Watanabe is an Associate Professor at Carnegie Mellon University in Pittsburgh and a leading researcher in speech and language processing. His work spans automatic speech recognition, speech enhancement, spoken language understanding, and machine learning for speech and language processing. He has contributed more than 500 publications to peer-reviewed journals and received several awards, including the best paper award from ISCA Interspeech 2024. In this episode, Associate Prof. Watanabe reflects on the transformative progress in speech recognition over the past decade, highlighting milestones from the adoption of deep neural networks to the rise of large-scale models like OpenAI Whisper. He discusses the ongoing challenges in achieving human-level understanding in complex scenarios such as multi-speaker conversations, accented and multilingual speech, and child or disordered speech. He concludes with thoughts on academia’s enduring role in shaping the field, and how his inspiration is often drawn from science fiction and Japanese animation.

In this episode of the IEEE Signal Processing Society podcast, Prof. Samson Lasaulce, Chief Research Scientist at Khalifa University (KU) interviews Prof. Merouane Debbah, founding Director of the KU 6G Research Center. They talk about how AI is transforming the future of wireless communications, the use of large language models (LLMs) to revolutionize network management, improve communication protocols, and advance automation. About the speaker Mérouane Debbah is a Professor at Khalifa University of Science and Technology in Abu Dhabi and founding Director of the KU 6G Research Center. His research bridges mathematics, algorithms, statistics, and communication sciences, with a focus on random matrix theory and learning algorithms. He has played a pivotal role in advancing small cells (4G), Massive MIMO (5G), and Large Intelligent Surfaces (6G) technologies. He is a frequent keynote speaker at international events in the field of telecommunication and AI. In this episode, Professor Debbah explores the transformative potential of LLMs in wireless communication. Key applications include dynamic spectrum management, end-to-end system optimization, fault detection, mobility management, and AI-based localization. These techniques enhance data handling and system performance by replacing complex modeling with efficient data-driven approaches. These insights shed light on how 6G, powered by pervasive AI, will revolutionize wireless networks and redefine communication frameworks for the future.

In this episode of the IEEE Signal Processing Society podcast, Dennis K. Chrogony, Education Board Outreach and Visibility Committee Member interviews Sergios Theodoridis, Professor Emeritus, Signal Processing and Machine Learning, National and Kapodistrian University of Athens, Greece, Aalborg University, Denmark, and Shenzhen Research Institute of Big Data, Chinese University of Hong Kong, China. They delve into the evolution and impact of signal processing and AI and machine learning on technological advancements. Professor Sergios Theodoridis Prof. Theodoridis has made countless contributions to the field of signal processing and machine learning. He has authored several books, received multiple awards, and earned prestigious accolades like the EURASIP Athanasios Papoulis Award and the IEEE Signal Processing Society Education Award. His leadership positions, including his role as Vice President of the IEEE Signal Processing Society highlight his significant influence and respect within the academic and professional community. In this episode, Prof. Theodoridis discusses the advancements in various areas of signal processing, highlighting AI and machine learning as pivotal technologies in modern society. He explains how the seamless integration of machine learning into signal processing has led to significant improvements in areas like speech and audio recognition and natural language processing. He also notes that while the core goals of signal processing remain unchanged, new techniques from the machine learning community have greatly enhanced its applications.

In the first podcast sponsored by the IEEE Signal Processing Society, Ervin Sejdić, Professor at University of Toronto’s Edward S. Rogers Sr. Department of Electrical & Computer Engineering interviews April Khademi, Associate Professor of Biomedical Computer and Electrical Engineering at Toronto Metropolitan University and Canada Research Chair in AI for Medical Imaging. April Khademi April, an expert in AI-driven medical image analysis, focuses on AI-driven signal processing methods that have revolutionized medical imaging, improving diagnostic and therapeutic accuracy. In this podcast, she discusses how AI advancements continuously push the boundaries of medical imaging, providing clinicians with robust, quantitative disease metrics and enhancing overall healthcare quality. AI-driven foundation models in medical imaging, like MRI and CT, enhance object detection and segmentation. They mitigate data constraints and enable fine-tuning for smaller datasets. FDA-approved algorithms in image acquisition improve efficiency, allowing faster, lower-dose scans and super-resolution for better image quality, yielding substantial business benefits. AI also enhances interrater agreement in medical image interpretation, reducing subjectivity and varying expertise among clinicians. It ensures consistent, accurate diagnoses, especially in community hospitals without specialized pathologists, ultimately leading to more reliable patient treatment outcomes.