)
🎙️ Are synthetic speech artifacts evenly distributed across phonemes? Not quite!

In our latest paper in MTAP, we propose a phonetic-driven framework using PPGs & phoneme pruning, boosting anti-spoofing performance on ASVspoof 2019 LA.

🔗 link.springer.com/article/10.1...

#SpeechAI #AudioDeepfake

🎙️ Does knowing a speaker’s gender actually boost speaker recognition?

In our latest paper in Soft Computing (Springer), we explore gender effects via bio-inspired filterbanks (Gammatone, Cascade, etc.)

🔗 link.springer.com/article/10.1...

#SpeechProcessing #AudioAI #AudioDeepFake #ISPlab

In our new paper in IEEE Access, we use Log-Area Ratios (LARs) + a novel Conditional Speaker Normalization (CSN) conditioned on speaker proxies (e.g. height) to detect synthetic speech reliably on ASVspoof & FoR.

ieeexplore.ieee.org/abstract/doc...

#DeepfakeDetection #SpeechAI #AudioDeepfake

A forensic similarity method can tell if two speech clips share the same manipulation traces, using a shallow similarity network—without large labelled data (Oct 3 2025). Read more: https://getnews.me/forensic-similarity-technique-advances-speech-deepfake-detection/ #audiodeepfake #forensics

AUDDT, an open‑source toolkit, combines 28 public audio deepfake datasets for benchmarking pretrained detectors; initial tests show accuracy drops on out‑of‑domain data. Read more: https://getnews.me/auddt-open-source-toolkit-for-benchmarking-audio-deepfake-detection/ #auddt #audiodeepfake

A new framework detects zero‑day audio deepfakes with retrieval‑based voice profiling, achieving performance on par with fine‑tuned models on DeepFake‑Eval‑2024. Read more: https://getnews.me/zero-day-audio-deepfake-detection-with-retrieval-and-profile-matching/ #audiodeepfake #zeroday

The new LAVA framework for voice attribution reports ADA F1 scores above 95% and ADMR macro F1 of 96.31% across six model classes, with confidence‑based rejection for unknown attacks. https://getnews.me/lava-framework-boosts-audio-deepfake-attribution-and-model-recognition/ #lavaa #audiodeepfake

Audio-Deepfakes – zunehmende KI-Verfügbarkeit lässt Betrugsversuche rasant ansteigen

#AudioDeepfake #Betrug #Deepfake #HumanRiskManagement #KITool @KnowBe4 #SecurityAwareness #Sicherheitsbewusstsein #Sprachanrufe

netzpalaver.de/2025/...