Digger - deepfake detection (Round 6)
Summary
When we cannot trust our eyes and ears anymore then what can we trust?
Deepfakes are hyper-realistic video forgeries in which people say or do things that they’ve never actually said or done. The visual quality of deepfakes will soon become so flawless that it will be hard to make a judgement on veracity by mere visual verification.
Digger’s solution is to approach the problem from a different side; using state-of-the-art audio forensics technologies to detect audio tampering in videos. The audio forensics toolkit will be integrated in a video verification application that is made available to verification experts.
The solution
Deepfakes are a real threat to society. Mere visual verification is not enough to make a judgement on veracity. Current technologies to check if footage has been altered are not reliable. Digger’s solution is to make use of state-of-the-art audio forensics technologies to develop a toolkit that can detect audio tampering in videos. This toolkit will be integrated in a video verification application that is made available to verification experts for integration in their workflows. The aim is to develop a community in order to share knowledge and initiate collaboration in the field of deepfake detection.