Few-shot adaptation with a self-attentive prototypical network reduces spoofed-speech detection errors by up to 32% relative under distribution shifts.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Rapidly Adapting to New Voice Spoofing: Few-Shot Detection of Synthesized Speech Under Distribution Shifts
Few-shot adaptation with a self-attentive prototypical network reduces spoofed-speech detection errors by up to 32% relative under distribution shifts.