CopyPaste-based parallel data, cosine similarity loss, and energy-based emotion masking reduce speaker verification EER by 19.29% relative on the Dusha emotional speech corpus.
Performance of the baseline system The performance of the pre-trained model on V oxCeleb1 is presented in Table 3, with the results of the ECAPA model referenced from [3]
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.SD 1years
2025 1verdicts
CONDITIONAL 1roles
baseline 1polarities
baseline 1representative citing papers
citing papers explorer
-
Learning Emotion-Invariant Speaker Representations for Speaker Verification
CopyPaste-based parallel data, cosine similarity loss, and energy-based emotion masking reduce speaker verification EER by 19.29% relative on the Dusha emotional speech corpus.