Speaker: Sara Barahona Quirós. Abstract: Cross-lingual speaker verification remains a challenging problem due to variability in language, recording conditions, and speaker characteristics. In this talk, we present our participation in the new benchmark TidyVoice Challenge: Cross-Lingual Speaker Verification, associated with… Read More
An Introduction to Ethics, Risks and Safety in Artificial Intelligence
Speaker: Joaquín González Rodríguez. Abstract: With the omnipresence of AI in the media, it has become increasingly difficult to separate relevant information from background noise, with predictions ranging from utopian promises to apocalyptic scenarios. Moreover, much of the existing scientific… Read More
Web Pentesting: How and Why Modern Applications Break
Speaker: Pablo González Escribano. Abstract: This talk introduces web pentesting, how it is performed, and why it remains relevant in modern applications. It discusses common assumptions about web security, professional pentesting methodologies, and analyzes real-world vulnerabilities such as IDOR, SQL… Read More
Summary of the CIFE 2026 (Congreso Internacional de Fonética Experimental)
Speaker: Manuel Otero González. Abstract: This presentation explained the development of the CIFE 2026 (Congreso Internacional de Fonética Experimental) and provided an overview of the various sessions attended. Some of these sessions explored the content presented and the lines of… Read More
A Perceptually-Optimised Hybrid Acoustic Simulation for Interactive Environments
Speaker: Maya Tia Kanani. Abstract: Accurate simulation of sound propagation is an ongoing problem. Current approaches are computationally demanding, especially in virtual environments where it is often necessary to consider complex geometries, real-time rendering and dynamic scenes. Whilst existing hybrid… Read More
Transformer Embeddings Enable Accurate Forensic Classification of Soil DNA
Speaker: Manuel Fernando Mollón Laorca. Abstract: In this study, we use transformer-based DNA embeddings from a pre-trained DNABERT-2 model to train a classifier that distinguishes soil evidence containing blood, urine, feces, cadaver material, or no biological material (control). We systematically… Read More
Automatic Speech Recognition and Speaker Diarization in Spanish: Dialectal Varieties and Rural Speech
Speaker: Laura Herrera Alarcón Abstract: Audio databases provide essential linguistic information, capturing verbal content as well as sociolinguistic features. The COSER corpus (Audible Corpus of Spoken Rural Spanish) is introduced. In addition, training and test splits have been created to… Read More
Anatomy of a Deepfake: Types of Attacks and Detection Evaluations
Speaker: Manuel Otero González. Abstract: This talk will explore the world of deepfakes: what they are, how they work, and why they have become one of the most debated technologies in recent years. Starting with a brief personal introduction, we… Read More
Mic-e-mouse and the new AI-based Ciber Attacks
Speaker: Juan Antonio Gordillo Gayo. Abstract: The vast amount of data generated by individuals today has fueled the rise of AI-driven technologies that make our lives more convenient and efficient. However, these same tools also create unprecedented opportunities for hackers… Read More
Calibration of Multiclass Classifiers: A Wee Tutorial.
Speaker: Daniel Ramos Castro. Abstract: Probabilistic predictions are vital for decision-making in many applications of machine learning and AI, including medicine, forensics, security, and safety. However, many multiclass classifiers produce poorly calibrated outputs, leading to suboptimal decisions with potentially high… Read More