PartialSpoof

The 'PartialSpoof' dataset is used to evaluate the detection of deepfake speech, specifically focusing on the challenges posed by partially edited audio generated through neural speech editing techniques.

Papers using PartialSpoof (5)

Split and Conquer Partial Deepfake Speech2026Frame-level Temporal Difference Learning for Partial Deepfake Speech Detection2025PartialEdit: Identifying Partial Deepfakes in the Era of Neural Speech Editing2025Nes2Net: A Lightweight Nested Architecture for Foundation Model Driven Speech Anti-spoofing2025The PartialSpoof Database and Countermeasures for the Detection of Short Fake Speech Segments Embedded in an Utterance2022