Almanac
paper

From Self-Supervised Speech Models to Mixture-of-Experts for Robust Anti-Spoofing

paperactiveprovisionalfrom-self-supervised-speech-models-to-mixture-of-experts-for-robust-anti-spoofing-817d69b8·1 events·first seen 2d ago

Aliases: From Self-Supervised Speech Models to Mixture-of-Experts for Robust Anti-Spoofing

Co-occurring entities

More like this (12)

Recent events (1)

3arXiv · cs.AI·2d ago·source ↗

MoE architecture improves self-supervised speech model robustness for anti-spoofing

Researchers propose converting a self-supervised speech representation model into a Mixture-of-Experts (MoE) architecture to improve generalization in synthetic speech detection. Feed-forward blocks in selected encoder layers are replaced by expert networks with a layer-wise gating mechanism, allowing complementary acoustic pattern capture while preserving pretrained representations. Evaluated across 14 spoofing datasets, the approach reduces macro Equal Error Rate from 5.46% to 4.81%, an 11.9% relative improvement over the baseline.