ISCA Archive Interspeech 2021
ISCA Archive Interspeech 2021

Bidirectional Multiscale Feature Aggregation for Speaker Verification

Jiajun Qi, Wu Guo, Bin Gu

In this paper, we propose a novel bidirectional multiscale feature aggregation (BMFA) network with attentional fusion modules for text-independent speaker verification. The feature maps from different stages of the backbone network are iteratively combined and refined in both a bottom-up and top-down manner. Furthermore, instead of simple concatenation or elementwise addition of feature maps from different stages, an attentional fusion module is designed to compute the fusion weights. Experiments are conducted on the NIST SRE16 and VoxCeleb1 datasets. The experimental results demonstrate the effectiveness of the bidirectional aggregation strategy and show that the proposed attentional fusion module can further improve the performance.


doi: 10.21437/Interspeech.2021-111

Cite as: Qi, J., Guo, W., Gu, B. (2021) Bidirectional Multiscale Feature Aggregation for Speaker Verification. Proc. Interspeech 2021, 71-75, doi: 10.21437/Interspeech.2021-111

@inproceedings{qi21_interspeech,
  author={Jiajun Qi and Wu Guo and Bin Gu},
  title={{Bidirectional Multiscale Feature Aggregation for Speaker Verification}},
  year=2021,
  booktitle={Proc. Interspeech 2021},
  pages={71--75},
  doi={10.21437/Interspeech.2021-111}
}