Search papers, labs, and topics across Lattice.
5
0
5
5
Current speech deepfake detection systems falter dramatically against emotionally expressive attacks, with performance dropping to near-random levels on the new AffectDF benchmark.
Even anonymized speech reveals over 15% more speaker information when combining audio and text, challenging current privacy assumptions.
Training on real speech prosody alone can cut speech deepfake error rates by over 70% on emotional attacks, a blindspot for current detectors.
LLMs can spot fake words in speech by recognizing common editing patterns, but this reliance on learned biases hinders generalization to new manipulation techniques.
Achieve zero-shot voice conversion competitive with methods requiring more data or training, using a simple, invertible linear method to disentangle speech content from speaker timbre.
