A New Perspective on Speaker Verification: Joint Modeling with DFSMN and Transformer

TL;DR

Voice Transformer (VOT) is proposed, a novel model for speaker verification, which integrates parallel transformers at multiple scales and incorporates a deep feedforward sequential memory network (DFSMN) into the attention part of these transformers to increase feature granularity.

Abstract

Speaker verification is to judge the similarity between two unknown voices in an open set, where the ideal speaker embedding should be able to condense discriminant information into a compact utterance-level representation that has small intra-speaker distances and large inter-speaker distances. We propose Voice Transformer (VOT), a novel model for speaker verification, which integrates parallel transformers at multiple scales. A deep feedforward sequential memory network (DFSMN) is incorporated into the attention part of these transformers to increase feature granularity. The attentive statistics pooling layer is added to focus on important frames and form utterance-level features. We propose Additive Angular Margin Focal Loss (AAMF) to address the hard samples problem. We evaluate the proposed approach on the VoxCeleb1 and CN-Celeb2 datasets, demonstrating that VOT surpasses most mainstream models. The code is available on GitHub\footnote{\url{https://github.com/luckyerr/Voice-Transformer_Speaker-Verification}}.

Figures (4)

• Figure 1: The Voice-Transformer architecture
• Figure 2: The architecture of DFSMN
• Figure 3: serial and parallel transformer structure architecture
• Figure 4: Comparison of features