In cross-lingual non-target speaker verification, comparisons involving languages seen in training cluster tightly regardless of comparison language, while comparisons among untrained languages produce scores that correlate with language similarity, especially under multilingual training.
A common research paradigm to address this question has been to study mismatch between the language spoken by the questioned speaker and known speaker [1, 2, 3, 4]
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
eess.AS 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
On the influence of language similarity in non-target speaker verification trials
In cross-lingual non-target speaker verification, comparisons involving languages seen in training cluster tightly regardless of comparison language, while comparisons among untrained languages produce scores that correlate with language similarity, especially under multilingual training.