# DAME: Duration-Aware Matryoshka Embedding for Duration-Robust Speaker Verification

arXiv: https://arxiv.org/abs/2601.13999 · HTML: https://mlsanity.com/paper/2601.13999/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 23 references that are in the map.
- First posted on arXiv 20 Jan 2026.

## Builds on (most cited first, 18 of 18)

- [VoxCeleb2: Deep Speaker Recognition](https://mlsanity.com/paper/1806.05622/) (2018)
- [VoxCeleb: a large-scale speaker identification dataset](https://mlsanity.com/paper/1706.08612/) (2017)
- [Cyclical Learning Rates for Training Neural Networks](https://mlsanity.com/paper/1506.01186/) (2015)
- [ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification](https://mlsanity.com/paper/2005.07143/) (2020)
- [MUSAN: A Music, Speech, and Noise Corpus](https://mlsanity.com/paper/1510.08484/) (2015)
- [Matryoshka Representation Learning](https://mlsanity.com/paper/2205.13147/) (2022)
- [In defence of metric learning for speaker recognition](https://mlsanity.com/paper/2003.11982/) (2020)
- [Wespeaker: A Research and Production oriented Speaker Embedding Learning Toolkit](https://mlsanity.com/paper/2210.17016/) (2022)
- [An Enhanced Res2Net with Local and Global Feature Fusion for Speaker Verification](https://mlsanity.com/paper/2305.12838/) (2023)
- [The IDLAB VoxSRC-20 Submission: Large Margin Fine-Tuning and Quality-Aware Score Calibration in DNN Based Speaker Verification](https://mlsanity.com/paper/2010.11255/) (2020)
- [ECAPA2: A Hybrid Neural Network Architecture and Training Strategy for Robust Speaker Embeddings](https://mlsanity.com/paper/2401.08342/) (2024)
- [ERes2NetV2: Boosting Short-Duration Speaker Verification Performance with Computational Efficiency](https://mlsanity.com/paper/2406.02167/) (2024)
- [The VOiCES from a Distance Challenge 2019 Evaluation Plan](https://mlsanity.com/paper/1902.10828/) (2019)
- [Frequency and Multi-Scale Selective Kernel Attention for Speaker Verification](https://mlsanity.com/paper/2204.01005/) (2022)
- [Meta-Learning for Short Utterance Speaker Recognition with Imbalance Length Pairs](https://mlsanity.com/paper/2004.02863/) (2020)
- [Exploring Binary Classification Loss For Speaker Verification](https://mlsanity.com/paper/2307.08205/) (2023)
- [Improving Multi-Scale Aggregation Using Feature Pyramid Module for Robust Speaker Verification of Variable-Duration Utterances](https://mlsanity.com/paper/2004.03194/) (2020)
- [M-Vec: Matryoshka Speaker Embeddings with Flexible Dimensions](https://mlsanity.com/paper/2409.15782/) (2024)

## Abstract (arXiv)

Short-utterance speaker verification remains challenging due to limited speaker-discriminative cues in short speech segments. While existing methods focus on enhancing speaker encoders, the embedding learning strategy still forces a single fixed-dimensional representation reused for utterances of any length, leaving capacity misaligned with the information available at different durations. We propose Duration-Aware Matryoshka Embedding (DAME), a model-agnostic framework that builds a nested hierarchy of sub-embeddings aligned to utterance durations: lower-dimensional representations capture compact speaker traits from short utterances, while higher dimensions encode richer details from longer speech. DAME supports both training from scratch and fine-tuning, and serves as a direct alternative to conventional large-margin fine-tuning, consistently improving performance across durations. On the VoxCeleb1-O/E/H and VOiCES evaluation sets, DAME consistently reduces the equal error rate on 1-s and other short-duration trials, while maintaining full-length performance with no additional inference cost. These gains generalize across various speaker encoder architectures under both general training and fine-tuning setups.