LASPA uses prefix-tuned cross-attention and multiple losses to remove language information from speaker embeddings, lowering error rates on multilingual speaker verification tests.
Our method effectively separates speaker iden- tity from linguistic information, addressing key challenges in multi-lingual speaker recognition
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.SD 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
LASPA: Language Agnostic Speaker Disentanglement with Prefix-Tuned Cross-Attention
LASPA uses prefix-tuned cross-attention and multiple losses to remove language information from speaker embeddings, lowering error rates on multilingual speaker verification tests.