Resources

Interactive demos, code, datasets, and project pages from ASPIRE and collaborators.

Try It Yourself

• Agent Security Playground Launch attacks on a sandboxed LLM agent
• EchoGuard Ultrasound human-presence verification
• SoK: How Robust is Audio Watermarking in Generative AI models? (arXiv 2025) Benchmark with listenable audio demos
• WiSenseHub Wireless-sensing dataset catalog
• Learning to Evade: Adaptive Attacks on Audio Watermarking (Interspeech 2026) Adaptive attacks on audio watermarking

Human–Mobile Security

• WavePurifier: Purifying Audio Adversarial Examples via Hierarchical Diffusion Models (MobiCom 2024) Demo · Code
• MASTERKEY: Practical Backdoor Attack Against Speaker Verification Systems (MobiCom 2023) Demo
• SpecPatch: Human-In-The-Loop Adversarial Audio Spectrogram Patch Attack on Speech Recognition (CCS 2022) Demo
• PhantomSound: Black-Box, Query-Efficient Audio Adversarial Attack via Split-Second Phoneme Injection (RAID 2023) Paper
• SurfingAttack: Interactive Hidden Attack on Voice Assistants Using Ultrasonic Guided Waves (NDSS 2021) Demo
• GhostTalk: Interactive Attack on Smartphone Voice System Through Power Line (NDSS 2022) Demo
• SuperVoice: Text-Independent Speaker Verification Using Ultrasound Energy in Human Speech (AsiaCCS 2022) Demo
• PiezoBud: A Piezo-Aided Secure Earbud with Practical Speaker Authentication (SenSys 2024) Paper

Human–Human Security

• Learning to Evade: Adaptive Attacks on Audio Watermarking (Interspeech 2026) Demo · Paper
• AUDIO WATERMARK: Dynamic and Harmless Watermark for Black-box Voice Dataset Copyright Protection (USENIX Security 2025) Demo · Code
• SoK: How Robust is Audio Watermarking in Generative AI models? (arXiv 2025) Demo · Project
• ClearMask: Noise-Free and Naturalness-Preserving Protection Against Voice Deepfake Attacks (AsiaCCS 2025) Demo
• NEC: Speaker Selective Cancellation via Neural Enhanced Ultrasound Shadowing (DSN 2022) Demo
• VSMask: Defending Against Voice Synthesis Attack via Real-Time Predictive Perturbation (WiSec 2023) Paper
• Secure-IRS: Defending Against Adversarial Physical-Layer Sensing in ISAC System (ICNC 2025) Demo · Paper
• A Survey on Reconfigurable Intelligent Surfaces in Practical Systems: Security and Privacy Perspectives (arXiv 2025) Curated reading list

Human–Agent Security

• Agent Security Playground Interactive playground
• EchoGuard Demo
• WebCloak: Characterizing and Mitigating Threats from LLM-Driven Web Agents as Intelligent Scrapers (IEEE S&P 2026) Demo
• On the Risks of Generative Engine Optimization in the Era of LLMs (ICML 2026) Demo · Paper
• Mental Damage: Caption Poisoning Attacks on Retrieval-Augmented Text-to-Music Generation (IEEE S&P Workshop ArtSec 2026) Demo
• Audio Jailbreak Attacks: Exposing Vulnerabilities in SpeechGPT in a White-Box Framework (DSN-DSML 2025) Code

Datasets & Systems

• WiSenseHub Dataset catalog
• NELoRa: Towards Ultra-low SNR LoRa Communication with Neural-enhanced Demodulation (SenSys 2021) Code