We present results and analyses from the third VoicePrivacy Challenge held in 2024, which focuses on advancing voice anonymization technologies. The task was to develop a voice anonymization system for speech data that conceals a speaker's voice identity while preserving linguistic content and emotional state. We provide a systematic overview of the challenge framework, including detailed descriptions of the anonymization task and datasets used for both system development and evaluation. We outline the attack model and objective evaluation metrics for assessing privacy protection (concealing speaker voice identity) and utility (content and emotional state preservation). We describe six baseline anonymization systems and summarize the innovative approaches developed by challenge participants. Finally, we provide key insights and observations to guide the design of future VoicePrivacy challenges and identify promising directions for voice anonymization research.
Future 6G radio access networks (RANs) will be artificial intelligence (AI)-native: observed, reasoned about, and re-configured by autonomous agents cooperating across the cloud-edge continuum. We introduce MX-AI, the first end-to-end agentic system that (i) instruments a live 5G Open RAN testbed based on OpenAirInterface (OAI) and FlexRIC, (ii) deploys a graph of Large-Language-Model (LLM)-powered agents inside the Service Management and Orchestration (SMO) layer, and (iii) exposes both observability and control functions for 6G RAN resources through natural-language intents. On 50 realistic operational queries, MX-AI attains a mean answer quality of 4.1/5.0 and 100
Task-oriented semantic communication (ToSC) emerges as an innovative approach in the 6G landscape, characterized by the transmission of only vital information that is directly pertinent to a specific task. While ToSC offers an efficient mode of communication, it concurrently raises concerns regarding privacy, as sophisticated adversaries might possess the capability to reconstruct the original data from the transmitted features. This paper provides an in-depth analysis of privacy-preserving strategies specifically designed for ToSC relying on deep neural network-based joint source and channel coding (DeepJSCC). Our study encompasses a detailed comparative assessment of trustworthy feature perturbation methods such as differential privacy (DP) and encryption, alongside intrinsic security incorporation approaches like adversarial learning to train the JSCC and learning-based vector quantization (LBVQ). Our comparative analysis underscores the integration of advanced explainable learning algorithms into communication systems, positing a new benchmark for privacy standards in the forthcoming 6G era.
Localization is increasingly becoming an integral component of wireless cellular networks. The advent of artificial intelligence (AI) and machine learning (ML) based localization algorithms presents potential for enhancing localization accuracy. Nevertheless, current standardization efforts in the third generation partnership project (3GPP) and the O-RAN Alliance do not support AI/ML-based localization. In order to close this standardization gap, this paper describes an O-RAN framework that enables the integration of AI/ML-based localization algorithms for real-time deployments and testing. Specifically, our framework includes an O-RAN E2 Service Model (E2SM) and the corresponding radio access network (RAN) function, which exposes the Uplink Sounding Reference Signal (UL-SRS) channel estimates from the E2 agent to the Near real-time RAN Intelligent Controller (Near-RT RIC). Moreover, our framework includes, as an example, a real-time localization external application (xApp), which leverages the custom E2SM-SRS in order to execute continuous inference on a trained Channel Charting (CC) model, which is an emerging self-supervised method for radio-based localization. Our framework is implemented with OpenAirInterface (OAI) and FlexRIC, democratizing access to AI-driven positioning research and fostering collaboration. Furthermore, we validate our approach with the CC xApp in real-world conditions using an O-RAN based localization testbed at EURECOM. The results demonstrate the feasibility of our framework in enabling real-time AI/ML localization and show the potential of O-RAN in empowering positioning use cases for next-generation AI-native networks.
Spoofing detection systems are typically trained using diverse recordings from multiple speakers, often assuming that the resulting embeddings are independent of speaker identity. However, this assumption remains unverified. In this paper, we investigate the impact of speaker information on spoofing detection systems. We propose two approaches within our Speaker-Invariant Multi-Task framework, one that models speaker identity within the embeddings and another that removes it. SInMT integrates multi-task learning for joint speaker recognition and spoofing detection, incorporating a gradient reversal layer. Evaluated using four datasets, our speaker-invariant model reduces the average equal error rate by 17
