Deep Speaker Embedding Extraction with Channel-Wise Feature Responses and Additive Supervision Softmax Loss Function

By Jianfeng Zhou, Tao Jiang, Zheng Li, Lin Li, Qingyang Hong. 2019

70 citations recorded by the source corpus.

In speaker verification, the convolutional neural networks (CNN) have been successfully leveraged to achieve a great performance. Most of the models based on CNN primarily focus on learning the distinctive speaker embedding from the horizontal direction (time-axis). However, the feature relationship between channels is usually neglected. In this paper, we firstly aim toward an alternate direction of recalibrating the channelwise features by introducing the recently proposed “squeezeand-excitation” (SE) module for image classification. We effectively incorporate the SE blocks in the deep residual networks (ResNet-SE) and demonstrate a slightly improvement on VoxCeleb corpuses. Additionally, we propose a new loss function, namely additive supervision softmax (AS-Softmax), to make full use of the prior knowledge of the mis-classified samples at training stage by imposing more penalty on the mis-classified samples to regularize the training process. The experimental results on VoxCeleb corpuses demonstrate that the proposed loss could further improve the performance of speaker system, especially on the case that the combination of the ResNet-SE and the AS-Softmax.

Subfields: Artificial Intelligence, Representation & Self-Supervised Learning, Speech & Audio, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Representation & Self-Supervised Learning

In Speech & Audio

In Signal Processing

Citations in the corpus

VOXLINGUA107: A Dataset for Spoken Language Recognition (2020, methodology): “For frame-level feature extraction, we use the Resnet34 [15, 16] architecture where the basic convolutional blocks with residual connections are replaced with squeeze-and-attention modules [17, 18].”
Speaker recognition based on deep learning: An overview (2020, methodology): “ns. Some work directly used the standard ResNet as their speaker feature extractors [68, 58, 59, 69]. Some other work employed ResNet as a backbone and modiﬁed it for speciﬁc purposes or applications [60, 92, 80, 61, 70, 71]. For example, to reduce the number of parameters, [60] modiﬁed the standard”
ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification (2020, methodology): “ule described above. For this purpose we introduce 1- dimensional Squeeze-Excitation (SE) blocks, as this computer vision approach to model global channel interdependencies has been proved successful [14, 15]. The ﬁrst component of an SE-block is the squeeze operation which generates a descriptor fo”