An iterative framework for self-supervised deep speaker representation learning
This work addresses speaker verification for audio processing, but it is incremental as it builds on existing self-supervised and clustering methods.
The paper tackled the problem of self-supervised speaker representation learning by proposing an iterative framework that uses clustering and pseudo-labeling to bootstrap a deep neural network, resulting in a 61% performance gain over a baseline contrastive loss model after 5 iterations on the VoxCeleb dataset.
In this paper, we propose an iterative framework for self-supervised speaker representation learning based on a deep neural network (DNN). The framework starts with training a self-supervision speaker embedding network by maximizing agreement between different segments within an utterance via a contrastive loss. Taking advantage of DNN's ability to learn from data with label noise, we propose to cluster the speaker embedding obtained from the previous speaker network and use the subsequent class assignments as pseudo labels to train a new DNN. Moreover, we iteratively train the speaker network with pseudo labels generated from the previous step to bootstrap the discriminative power of a DNN. Speaker verification experiments are conducted on the VoxCeleb dataset. The results show that our proposed iterative self-supervised learning framework outperformed previous works using self-supervision. The speaker network after 5 iterations obtains a 61% performance gain over the speaker embedding model trained with contrastive loss.