# ERes2NetV2: Boosting Short-Duration Speaker Verification Performance with Computational Efficiency

arXiv: https://arxiv.org/abs/2406.02167 · HTML: https://mlsanity.com/paper/2406.02167/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Jun 2024.

## Built on by (most cited first, 18 of 18)

- [Seeing, Listening, Remembering, and Reasoning: A Multimodal Agent with Long-Term Memory](https://mlsanity.com/paper/2508.09736/) (2025)
- [Nes2Net: A Lightweight Nested Architecture for Foundation Model Driven Speech Anti-spoofing](https://mlsanity.com/paper/2504.05657/) (2025)
- [SpeakerLM: End-to-End Versatile Speaker Diarization and Recognition with Multimodal Large Language Models](https://mlsanity.com/paper/2508.06372/) (2025)
- [StoryTeller: Improving Long Video Description through Global Audio-Visual Character Identification](https://mlsanity.com/paper/2411.07076/) (2024)
- [Towards Quantifying and Reducing Language Mismatch Effects in Cross-Lingual Speech Anti-Spoofing](https://mlsanity.com/paper/2409.08346/) (2024)
- [Malacopula: adversarial automatic speaker verification attacks using a neural-based generalised Hammerstein model](https://mlsanity.com/paper/2408.09300/) (2024)
- [The First VoicePrivacy Attacker Challenge](https://mlsanity.com/paper/2504.14183/) (2025)
- [Marco-Voice Technical Report](https://mlsanity.com/paper/2508.02038/) (2025)
- [Hermes the Polyglot: A Unified Framework to Enhance Expressiveness for Multimodal Interlingual Subtitling](https://mlsanity.com/paper/2602.00597/) (2026)
- [The First VoicePrivacy Attacker Challenge Evaluation Plan](https://mlsanity.com/paper/2410.07428/) (2024)
- [Identity as Presence: Towards Appearance and Voice Personalized Joint Audio-Video Generation](https://mlsanity.com/paper/2603.17889/) (2026)
- [Towards Lightweight and Stable Zero-shot TTS with Self-distilled Representation Disentanglement](https://mlsanity.com/paper/2501.08566/) (2025)
- [DAME: Duration-Aware Matryoshka Embedding for Duration-Robust Speaker Verification](https://mlsanity.com/paper/2601.13999/) (2026)
- [CineSRD: Leveraging Visual, Acoustic, and Linguistic Cues for Open-World Visual Media Speaker Diarization](https://mlsanity.com/paper/2603.16966/) (2026)
- [Reasoning LLM Improves Speaker Recognition in Long-form TV Dramas](https://mlsanity.com/paper/2607.02504/) (2026)
- [Don't Just Listen, Try Planning: Graph-based Retrieval-Generation Agent for Long-form Audio Meeting Understanding](https://mlsanity.com/paper/2608.24048/) (2026)
- [Not Quite My Tempo: Voice Activity-aware Speech Synthesis for Lip-Synchronous Dubbing](https://mlsanity.com/paper/2609.26486/) (2026)
- [CUE-Mem: Benchmarking Long-Term User Memory via Implicit Cues in Multimodal Conversations](https://mlsanity.com/paper/2609.32574/) (2026)

## Abstract (arXiv)

Speaker verification systems experience significant performance degradation when tasked with short-duration trial recordings. To address this challenge, a multi-scale feature fusion approach has been proposed to effectively capture speaker characteristics from short utterances. Constrained by the model's size, a robust backbone Enhanced Res2Net (ERes2Net) combining global and local feature fusion demonstrates sub-optimal performance in short-duration speaker verification. To further improve the short-duration feature extraction capability of ERes2Net, we expand the channel dimension within each stage. However, this modification also increases the number of model parameters and computational complexity. To alleviate this problem, we propose an improved ERes2NetV2 by pruning redundant structures, ultimately reducing both the model parameters and its computational cost. A range of experiments conducted on the VoxCeleb datasets exhibits the superiority of ERes2NetV2, which achieves EER of 0.61% for the full-duration trial, 0.98% for the 3s-duration trial, and 1.48% for the 2s-duration trial on VoxCeleb1-O, respectively.