StreamMark trains an Encoder-Distortion-Decoder network to embed semi-fragile watermarks that remain recoverable after benign audio transformations but drop to random accuracy under voice conversion and editing attacks.
Detecting voice cloning attacks via timbre watermarking
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
StreamMark: A Deep Learning-Based Semi-Fragile Audio Watermarking for Proactive Deepfake Detection
StreamMark trains an Encoder-Distortion-Decoder network to embed semi-fragile watermarks that remain recoverable after benign audio transformations but drop to random accuracy under voice conversion and editing attacks.