MECT: Mixture of Experts with CNN-Transformer Network for Speaker verification

arXiv:2609.24061 · cs.SD, cs.AI, eess.AS · Submitted 2026-09-21 · Read on arXiv

cs.SD, cs.AI, eess.AS

Submitted: 2026-09-21

Updated: 2026-09-21

Comments: 5 pages

Code: https://github.com/ant-research/AntSpeaker

License: http://creativecommons.org/licenses/by-nc-sa/4.0/

The gist: In this paper, we propose MECT, a speaker verification model that integrates the Mixture-of-Experts (MoE) mechanism into a CNN-Transformer backbone with optimized block structure and stacking scheme.

Terminology

Abstract

In this paper, we propose MECT, a speaker verification model that integrates the Mixture-of-Experts (MoE) mechanism into a CNN-Transformer backbone with optimized block structure and stacking scheme. Specifically, we investigated four MoE variants that span utterance-level and frame-level granularity with dense and sparse routing strategies. The MoE mechanism proves to be effective over the baseline without MoE with only a small increase in parameters. We further scale MECT to a series of model sizes, all maintaining compact parameters and low computational complexity. In particular, MECT-B2 achieves state-of-the-art performance on VoxCeleb1 and delivers strong results on CN-Celeb, demonstrating its effectiveness across diverse datasets. In addition, we establish a streaming inference paradigm through causal retraining, which maintains strong performance at a chunk size of 100ms.

Sources

Related papers