This multimodal dataset consists of recorded video interviews with identity-verified professionals who have provided consent for data use. It includes interview videos, conversation transcripts, anonymized CVs, liveness labels, and fraud-detection flags across engineering, finance, legal, and medical fields.
Potential Use Cases
- Deepfake and Synthetic Voice Detection Training: Authentic video and audio recordings from identity-verified individuals can serve as genuine reference data for training and evaluating models designed to distinguish real media from synthetic or manipulated content.
- Remote Identity Verification and Fraud Detection: Interview footage, liveness labels, and misconduct detection flags can support the analysis of suspicious activity and potential proxy interviews during remote identity verification.