Search papers, labs, and topics across Lattice.
Affiliation:
2
0
3
Achieving an unprecedented equal error rate of 0.88% in audio deepfake detection, this model sets a new benchmark for robustness against sophisticated voice synthesis attacks.
Language-specific statistical graphs reveal phoneme confusion patterns, achieving an F1-score of 59.52% in mispronunciation detection鈥攐utperforming existing methods.
