Multilingual speech foundation models, fused with a Tucker-Hadamard module, are reported to reach about 1% equal error rate on emotion fake audio detection, a large drop from prior benchmarks.
Speech Foundation Models (SFMs): For monolingual SFMs, we consider WavLM1 [17], Unispeech-SAT2 [18], Wav2vec23 [19], and HuBERT 4 [20]
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Enhancing In-Domain and Out-Domain EmoFake Detection via Cooperative Multilingual Speech Foundation Models
Multilingual speech foundation models, fused with a Tucker-Hadamard module, are reported to reach about 1% equal error rate on emotion fake audio detection, a large drop from prior benchmarks.