Pith. sign in

Then, we introduce the details of the multilingual soft content encoder, including the foundation models and fine-tuning procedures

Methods In this section, we describe framework of the proposed SSLbased multilingual SAS, which is illustrated in Figure 2

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

Mitigating Language Mismatch in SSL-Based Speaker Anonymization

eess.AS · 2025-07-01 · conditional · novelty 5.0

Fine-tuning an SSL content encoder on Japanese, especially when the encoder is pre-trained multilingually, makes anonymized Japanese and Mandarin speech much more intelligible while keeping speaker privacy at usable levels.

citing papers explorer

Showing 1 of 1 citing paper.

  • Mitigating Language Mismatch in SSL-Based Speaker Anonymization eess.AS · 2025-07-01 · conditional · none · ref 5

    Fine-tuning an SSL content encoder on Japanese, especially when the encoder is pre-trained multilingually, makes anonymized Japanese and Mandarin speech much more intelligible while keeping speaker privacy at usable levels.