NEWS
🎤 Confirmed Keynote Speaker: Prof. Elisabeth André
We are excited to confirm Prof. Elisabeth André as our second keynote speaker!!! 👏🏻👏🏻👏🏻
In her keynote, entitled “From Speech to Multimodal Interaction: Guardrails for Socially-Interactive AI”, Elisabeth will discuss how speech technologies evolve when moving from transcription to socially-aware multimodal interaction. The talk will explore the design of advanced guardrails for conversational AI systems, addressing uncertainty handling, policy compliance, reliability, and socially responsible interaction in complex real-world scenarios. The presentation will showcase insights from interdisciplinary projects involving social robots, virtual agents for children, healthcare assistants, and LLM-based in-car conversational systems.
About the Speaker
Elisabeth André is Full Professor of Computer Science and Founding Chair of Human-Centered Artificial Intelligence at Augsburg University, Germany. Internationally recognized for her pioneering work in multimodal human-machine interaction and socially intelligent AI, she has made foundational contributions to virtual agents and social robots capable of natural interaction through speech, gestures, and emotions.
We are honored to welcome Prof. Elisabeth André to our keynote program.
🎤 Confirmed Keynote Speaker: João Magalhães
We are delighted to announce João Magalhães as a confirmed keynote speaker!!! 👏🏻👏🏻👏🏻
In his talk, entitled “Multimodal Conversational Assistance of Complex Manual Tasks”, he will present TWIZ, a multimodal conversational AI assistant designed to guide users through complex real-world tasks. The keynote will highlight advances in human-like conversational design, multimodal interaction (voice, images, video), and robust zero-shot conversational flows.
About the Speaker
João Magalhães is Full Professor at Universidade NOVA de Lisboa and co-Director of the CMU Portugal partnership. His research focuses on multimodal AI, bridging vision and language for more human-like interaction. He has led numerous national and EU projects and received several distinctions, including 1st prize in the Amazon Alexa Taskbot Challenge 2022.
We look forward to his inspiring keynote!!!
ALBAYZIN 2026 Challenge: Tackling Dialectal Spanish with Speech-COSER
The ALBAYZIN 2026 Speech-COSER Challenge, part of IberSPEECH 2026, focuses on advancing speech technologies using the largest corpus of rural dialectal Spanish.
Participants can compete in ASR, speaker diarization, and spoken term detection, tackling one of the most diverse and challenging Spanish speech datasets.
📢 Call for Papers — IberSPEECH 2026
Join IberSPEECH 2026 in Madrid this November and share your latest advances in speech and language technologies. Paper submissions open until June 2026.
Take the Challenge: ALBAYZIN 2026 Evaluations
The ALBAYZIN Evaluations return to IberSPEECH 2026 with a three-track challenge focused on Automatic Speech Recognition, Speaker Diarization, and Spoken Term Detection using the COSER Spanish Rural Speech Corpus. Researchers and practitioners are invited to express their interest and take part in this benchmark initiative advancing Spanish speech technologies.
Support Innovation at IberSPEECH 2026!
We are excited to announce the Call for Sponsors for IberSPEECH 2026, the XIV Speech Technology Conference and the X Iberian SLTech Workshop. Taking place from November 18-20, 2026, in Madrid, this premier event serves as the ultimate meeting point for academic researchers, industry professionals, and innovation-driven institutions in the field of speech and language technologies.
GOLD SPONSOR
