Dimension-Specific Margins and Element-Wise Gradient Scaling for Enhanced Matryoshka Speaker Embedding

Mar 11, 2025·

Sunchan Park

,

Hyung Soon Kim

· 0 min read

Site

Speaker Recognition Representation Learning

Authors

Sunchan Park (he/him)

Postdoctoral Researcher

---

← Group-wise Layer Aggregation of Self-Supervised Representations for Speech Emotion Recognition Jun 23, 2026

Reverberant Environment Embedding using Dereverberation Autoencoder Jan 4, 2025 →