Pith. sign in

Additionally, it is better suited to our closed-set identi- fication task than other commonly used measures, such as equal error rate (EER)

Results We evaluated our models using balanced accuracy since it provides a more calibrated measure of performance than standard accuracy in the case of potential class imbalance

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
browse 1 citing papers

fields

eess.AS 1

years

2025 1

verdicts

CONDITIONAL 1

representative citing papers

Rhythm Features for Speaker Identification

eess.AS · 2025-06-07 · conditional · novelty 6.0

Using character-duration sequences from WhisperX, a transformer identifies speakers with balanced accuracy of 0.39 on LibriSpeech but only 0.03 on VoxCeleb1, and fusion with x-vectors does not improve accuracy.

citing papers explorer

Showing 1 of 1 citing paper.

  • Rhythm Features for Speaker Identification eess.AS · 2025-06-07 · conditional · none · ref 5

    Using character-duration sequences from WhisperX, a transformer identifies speakers with balanced accuracy of 0.39 on LibriSpeech but only 0.03 on VoxCeleb1, and fusion with x-vectors does not improve accuracy.