Large Margin Softmax Loss for Speaker Verification

Abstract

In neural network based speaker verification, speaker embedding is expected to be discriminative between speakers while the intra-speaker distance should remain small. A variety of loss functions have been proposed to achieve this goal. In this paper, we investigate the large margin softmax loss with different configurations in speaker verification. Ring loss and minimum hyperspherical energy criterion are introduced to further improve the performance. Results on VoxCeleb show that our best system outperforms the baseline approach by 15\% in EER, and by 13\%, 33\% in minDCF08 and minDCF10, respectively.

submitted to Interspeech 2019. The code and models have been released

References in corpus (1)

• Utterance-level Aggregation For Speaker Recognition In The Wild

Cited by in corpus (6)

• The SpeakIn System for VoxCeleb Speaker Recognition Challange 2021
• Text-Independent Speaker Verification with Dual Attention Network
• Adaptive Margin Circle Loss for Speaker Verification
• Contraction Mapping of Feature Norms for Classifier Learning on the Data with Different Quality
• AnyoneNet: Synchronized Speech and Talking Head Generation for Arbitrary Person
• Additive Phoneme-aware Margin Softmax Loss for Language Recognition