Speaker embeddings by modeling channel-wise correlations
This is an incremental improvement for speaker recognition systems, potentially enhancing embedding extraction in audio processing.
The paper tackled speaker recognition by proposing a new pooling method that uses pairwise correlations between channels for given frequencies as statistics, inspired by style-transfer in computer vision. Experiments on VoxCeleb demonstrated its effectiveness, though no concrete numbers were provided.
Speaker embeddings extracted with deep 2D convolutional neural networks are typically modeled as projections of first and second order statistics of channel-frequency pairs onto a linear layer, using either average or attentive pooling along the time axis. In this paper we examine an alternative pooling method, where pairwise correlations between channels for given frequencies are used as statistics. The method is inspired by style-transfer methods in computer vision, where the style of an image, modeled by the matrix of channel-wise correlations, is transferred to another image, in order to produce a new image having the style of the first and the content of the second. By drawing analogies between image style and speaker characteristics, and between image content and phonetic sequence, we explore the use of such channel-wise correlations features to train a ResNet architecture in an end-to-end fashion. Our experiments on VoxCeleb demonstrate the effectiveness of the proposed pooling method in speaker recognition.