Multi-Task Learning with High-Order Statistics for X-vector based Text-Independent Speaker Verification
This work addresses speaker verification for security or voice recognition applications, but it is incremental as it builds on existing x-vector methods.
The paper tackled improving speaker verification by proposing a multi-task learning architecture that combines speaker classification with reconstructing utterance statistics, resulting in outperforming the original x-vector approach on datasets like NIST SRE16 and VOiCES with low added complexity.
The x-vector based deep neural network (DNN) embedding systems have demonstrated effectiveness for text-independent speaker verification. This paper presents a multi-task learning architecture for training the speaker embedding DNN with the primary task of classifying the target speakers, and the auxiliary task of reconstructing the first- and higher-order statistics of the original input utterance. The proposed training strategy aggregates both the supervised and unsupervised learning into one framework to make the speaker embeddings more discriminative and robust. Experiments are carried out using the NIST SRE16 evaluation dataset and the VOiCES dataset. The results demonstrate that our proposed method outperforms the original x-vector approach with very low additional complexity added.