Additive Phoneme-aware Margin Softmax Loss for Language Recognition
This work addresses a specific bottleneck in language recognition by enhancing loss functions with phonetic information, representing an incremental improvement over existing methods.
The paper tackles the problem of suboptimal constant margins in softmax losses for language recognition by proposing an additive phoneme-aware margin softmax (APM-Softmax) loss, which automatically tunes margins based on phoneme recognition results, leading to improved performance over AM-Softmax and AAM-Softmax losses on Oriental Language Recognition datasets.
This paper proposes an additive phoneme-aware margin softmax (APM-Softmax) loss to train the multi-task learning network with phonetic information for language recognition. In additive margin softmax (AM-Softmax) loss, the margin is set as a constant during the entire training for all training samples, and that is a suboptimal method since the recognition difficulty varies in training samples. In additive angular margin softmax (AAM-Softmax) loss, the additional angular margin is set as a costant as well. In this paper, we propose an APM-Softmax loss for language recognition with phoneitc multi-task learning, in which the additive phoneme-aware margin is automatically tuned for different training samples. More specifically, the margin of language recognition is adjusted according to the results of phoneme recognition. Experiments are reported on Oriental Language Recognition (OLR) datasets, and the proposed method improves AM-Softmax loss and AAM-Softmax loss in different language recognition testing conditions.